• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

AMD Claims Arm ISA Doesn't Offer Efficiency Advantage Over x86

DukenukemX said:
ARM and X86 are nearly identical but ARM is a complete mess.
Click to expand...
messy by how similar to x86 it becamse (and x86 over time also completletly change once it started to translated evreything in uCode), a bit like how messy avx512 went and was removed with Intel... same on arm with their equivalent stuff.

x86 become over time pseudo RISC with the pentium, they do not execute variable lenght instruction on the hardware, they decode everything into ARM Like single uop and ARM has so many extension that it is not some limited set of simple core one.

ARM still easier to decode on the chip too.

By the end of the video, he ended up with about twice the efficacy of a steam deck and achieved to be gpu limited a lot of the time and took him only a few months too close the gap with x86 having many years dedicated to it.

DukenukemX said:
Also LLVM for ARM is really bad. So bad that he had to do it in assembly to fix some bugs.
Click to expand...
I doubt that very much, Apple is incredibly involved with LLVM (LLVM creator went to apple and they did finance a lot of the work) and they are really good with arm cpu, same with Google.

Having some issues come up when you do something like this:
The SPU recompiler is typically the heaviest part of emulating the PlayStation 3 on x86 machines. Despite the x86 back and having been relentlessly optimized, I dove straight into the assembly that LLVM produced. And this is what I saw.
Just as a recap, RPCS3 uses the LVM project to translate PlayStation 3 game code to x86 or ARM. We take PlayStation 3 assembly, translate that to LLVM, then LVM takes that and outputs native code for your machine. It's not dissimilar to how games like Unleash Reompiled work, only they're translating to C rather than LLVM IR. Unleash Recompiled even requires use of the LLVMbased Clang compiler to recompile the C code it produces.

That quite unusual and a high bar to use, I am sure there was a long list of issue and manual work involved and years going with the x86 version.

Sometime it did better for ARM, sometime it did better for x86, if you compare GCC with LLVM/clang result on arm you will not see some massive gap between the 2. Or more bugs/performance if you target x86 vs arm with LLVM/clang, specially not for Apple ARM cpus, the qualcomm used here with 4 different cores type was a bit of an hard case.
 
Last edited:
LukeTbk said:
By the end of the video, he ended up with about twice the efficacy of a steam deck and achieved to be gpu limited a lot of the time and took him only a few months too close the gap with x86 having many years dedicated to it.
Click to expand...
Keep in mind that most of the x86 optimizations he has already implemented were just adopted over to ARM, so of course it only took him a few months. The AYN Odin 2 is also using a more advanced 4nm manufacturing process for the Qualcomm chip. Also, that handheld is no longer $300.
I doubt that very much, Apple is incredibly involved with LLVM (LLVM creator went to apple and they did finance a lot of the work) and they are really good with arm cpu, same with Google.
Click to expand...
He mentions the problems with LLVM often. This part here is what I was thinking. I can't find the assembly solution part as the video is an hour long.

View: https://youtu.be/-aI_XEwmKFk?t=1636
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
DukenukemX said:
He mentions the problems with LLVM often. This part here is what I was thinking. I can't find the assembly solution part as the video is an hour long.
Click to expand...
Yes there is some problem with LLVM, but what he is doing is quite a special use case (just in time versus static compilation...), we do not know how many issues he had doing it on a x86 target, I am sure he had to do manual work has well.

3 of the biggest contributor to LLVM are Apple, Google and ARM, giant proportion of the world is running on compiled for ARM by LLVM (amazon graviton stuff, google arm cpu, all the mobile world) it is really not some second citizen/second thought.

MIcrosoft/Intel/Google do work hard on the x86 side of things and it is also really good has well.
 
LukeTbk said:
Yes there is some problem with LLVM, but what he is doing is quite a special use case (just in time versus static compilation...), we do not know how many issues he had doing it on a x86 target, I am sure he had to do manual work has well.

3 of the biggest contributor to LLVM are Apple, Google and ARM, giant proportion of the world is running on compiled for ARM by LLVM (amazon graviton stuff, google arm cpu, all the mobile world) it is really not some second citizen/second thought.

MIcrosoft/Intel/Google do work hard on the x86 side of things and it is also really good has well.
Click to expand...
I'm just pointing out how LLVM isn't entirely ready for ARM. I'm sure LLVM had problems on x86 as well. It's not the first time an emulator author has found bugs and reported them. Delroth had to work with MESA to get some bugs fixed for Dolphin.
 
Seems to be a secret x86 CPU being made that will make use of X86S. Said mysterious x86 CPU is said to have more advanced AMX instructions. X86 won't be a duopoly for very long.
https://www.phoronix.com/news/x86-CPU-With-APX-x86S

"I have been asked to document the x86 behavior for a PM64-only PML5-only implementation with FRED and APX that has been used at scale for >1 year.
...
After more than two decades of x86-64, Legacy Mode and Compatibility Mode fulfilled their purpose; it is time to drop the ballast which they pose.

Five years ago Intel published the first FRED spec and the dust for it settled in late 2023 and early 2024; Intel just shipped their first products with FRED, and AMD is expected to follow soon. The FRED cleanup for flexible return and event delivery was a crucial milestone in the evolution of x86.

Three years ago Intel published the first APX spec and the first x86S spec. For the APX spec the dust settled in late 2024 and early 2025; Intel and AMD are expected to finally ship their products in the not too distant future. On the other hand, for the x86S spec, Intel gave up in mid 2024. Which turned out to be premature. Someone else continued."
Click to expand...
 
Last edited:
Aurelius said:
That's all true, but that just illustrates the problem with AMD's rationale: put the right ingredients in place and ARM becomes incredibly efficient. Someone shopping for a portable AV editing workstation or just a quiet-but-quick productivity laptop isn't going to care about AMD's pitch. They're going to care that they can finish a video editing project entirely on battery power, or that they can write a report in blissful silence without invoking a special mode.
Click to expand...

Maybe I don't know the right people, but who the heck is doing an entire video editing project with no external monitors on battery? Is this a thing now? People sitting in coffee shops doing major tasks and expecting batteries to last that long? Most of the people I know who run their laptops on battery all day long are kids who are just too lazy to plug the thing in. They'll run it down to 5% while sitting next to a wall plug and never think to plug it in. I can see a very edge case for people doing heavy tasks on battery but most battery stuff is people waiting for a flight, sitting in a coffee shop catching up on some emails, or other such light tasks. Normal people (in my world) don't do major tasks while running around being mobile.
 
Ididar said:
Maybe I don't know the right people, but who the heck is doing an entire video editing project with no external monitors on battery? Is this a thing now? People sitting in coffee shops doing major tasks and expecting batteries to last that long? Most of the people I know who run their laptops on battery all day long are kids who are just too lazy to plug the thing in. They'll run it down to 5% while sitting next to a wall plug and never think to plug it in. I can see a very edge case for people doing heavy tasks on battery but most battery stuff is people waiting for a flight, sitting in a coffee shop catching up on some emails, or other such light tasks. Normal people (in my world) don't do major tasks while running around being mobile.
Click to expand...
You're quoting something that wasn't even true in 2025, let alone 2026. Both AMD and Intel machines can last all day on battery while video editing, and not lose much in performance. The exception is if the laptop has an RTX GPU which isn't great for battery life but is great for gaming. The fantasy for Apple users is to sit in a coffee shop to do work, because they're narcissists. Nobody in their right mind would do that unless they wanna make their lives more difficult. You sit at a desk with the laptop plugged in with an external mouse. The Mac touchpads are nice, but not nice enough to replace a $15 Logitech mouse.
 
DukenukemX said:
You're quoting something that wasn't even true in 2025, let alone 2026. Both AMD and Intel machines can last all day on battery while video editing, and not lose much in performance. The exception is if the laptop has an RTX GPU which isn't great for battery life but is great for gaming. The fantasy for Apple users is to sit in a coffee shop to do work, because they're narcissists. Nobody in their right mind would do that unless they wanna make their lives more difficult. You sit at a desk with the laptop plugged in with an external mouse. The Mac touchpads are nice, but not nice enough to replace a $15 Logitech mouse.
Click to expand...

This is why I challenged it. Laptop battery life at this stage just seems like bragging rights and not actually providing a demonstratable need in industry. My laptop at work might go on battery once a month when I'm dragged into a boardroom to do a presentation. Other than that it is plugged into the dock.
 
DukenukemX said:
Seems to be a secret x86 CPU being made that will make use of X86S. Said mysterious x86 CPU is said to have more advanced AMX instructions. X86 won't be a duopoly for very long.
https://www.phoronix.com/news/x86-CPU-With-APX-x86
Click to expand...

Wouldn't get your hopes up to much. This is either some robotics thing. China. Or it won't matter anyway.
X86 competition to AMD and Intel is never happening again.

The only way to do it legally at this point would be to extend some less then thing that was licensed but would be incomplete in many ways for general use. So hyper focused on a industry like robotics. OR it grows out of the Via license which has limitations and is also in Chinese hands.

The only chance there ever was, was transmeta. That failed cause the amount of money required to hang long enough to be a factor is insane. When transmeta failed all their IP was snapped up, and the most important bits that would allow a Transmeta V2 are now owned by Intel.

It would be nice to see competition in the X86 space, still never going to happen. Which is why things like ARM will continue to grow. Its not because anything is "better". There are pros and cons with both ISAs. The end of the day though Intel/AMD are not going to start handing out golden tickets to welcome competition on their core business. The only escapes are things like Via, and rosaiclabs that have old agreements... but are not current or complete enough to really compete for real either.
 
Ididar said:
This is why I challenged it. Laptop battery life at this stage just seems like bragging rights and not actually providing a demonstratable need in industry. My laptop at work might go on battery once a month when I'm dragged into a boardroom to do a presentation. Other than that it is plugged into the dock.
Click to expand...
Realistically, most laptops are used near an outlet but also realistically there's maybe an hour of battery life difference between Apple vs AMD and Intel. Apple still has a battery life advantage but it's marginal at best. The main advantage Apple has is performance because neither AMD nor Intel are building chips that compete with Apple's Max line of products. AMD did with Strix Halo and it isn't cheap, and it's also outdated in late 2026. The only reason AMD made Strix Halo is the same reason Nvidia created their DGX Spark, because of AI. Nvidia's RTX laptop GPU's are attractive enough that people don't care if AMD and Intel make a powerful monolithic chip, especially when it comes to CUDA.
ChadD said:
Wouldn't get your hopes up to much. This is either some robotics thing. China. Or it won't matter anyway.
X86 competition to AMD and Intel is never happening again.

The only way to do it legally at this point would be to extend some less then thing that was licensed but would be incomplete in many ways for general use. So hyper focused on a industry like robotics. OR it grows out of the Via license which has limitations and is also in Chinese hands.

The only chance there ever was, was transmeta. That failed cause the amount of money required to hang long enough to be a factor is insane. When transmeta failed all their IP was snapped up, and the most important bits that would allow a Transmeta V2 are now owned by Intel.
Click to expand...
I doubt someone made an x86 chip just to be used for robotics. Would make far more sense to use ARM. The only reason to build an x86 chip is for the massive amounts of software that already exist on it. Via's license is how I'd imagine they were able to legally build one.
It would be nice to see competition in the X86 space, still never going to happen. Which is why things like ARM will continue to grow. Its not because anything is "better". There are pros and cons with both ISAs. The end of the day though Intel/AMD are not going to start handing out golden tickets to welcome competition on their core business. The only escapes are things like Via, and rosaiclabs that have old agreements... but are not current or complete enough to really compete for real either.
Click to expand...
The reason x86 still holds a market is because it has decades worth of software that still work on modern processors. Something you can't say about ARM. Also, x86 has been 64-bit longer than ARM. Not to forget that Apple and Qualcomm are the only ones with fast ARM processors, and nobody else. Right now, you won't see ARM making inroads to desktop computers and the same goes for x86 to mobile devices. Unless ARM actually makes a design that's equivalent to Apple's, and AMD/Intel make their x86 designs small and efficient enough to be put in mobile devices.?
 
DukenukemX said:
I doubt someone made an x86 chip just to be used for robotics. Would make far more sense to use ARM. The only reason to build an x86 chip is for the massive amounts of software that already exist on it. Via's license is how I'd imagine they were able to legally build one.
Click to expand...

I am fairly sure that is exactly what it actually is.
Its not as crazy as it sounds. If you don't already own a ARM license you do have to buy it. If you already own a Atom license which they do.... I mean why not. Zero costs vs cost is pretty easy math. For a company that doesn't already have that old license in their back pocket sure, ARM all the way.

The bottom line is new companies can not get into x86. Its just legally impossible with out Intel and AMD inviting them in. Even with Intel they couldn't license modern full on x86 even if they wanted to. AMD likewise can't. The only way its possible is if both agree. Which is simply never ever going to happen. Never mind all the bottlenecks with SIMD instruction sets. There is basically ZERO legal way to create a x86 CPU, or X86 translation software without being dragged into law suits from Intel and likely AMD as well. Technically the transmeta patents for code morphing have expired I believe. However Transmeta was a wisker from loosing in court to Intel, the only thing that saved that for them was that Intel being lazy ass Intel stole their long run tech. So in a cross agreement Intel dropped their x86 suit (they never actually lost) and Transmeta granted Intel basically life long patents to all the Transmeta long run tech they stole. (basically the same script of how it went down with Cyrix)

We know something small scale is happening in x86 that isn't Intel and AMD. We also know it has to be something frankly that one or the both of them know about. The only two options that make any sense are the robotics type play based on licensed Atom arch. OR some chip based on the VIA patents, which means its going to lack basically anything Intel and AMD have done with x86 for the past 25 years. (there is a reason the Chinese x86 chips suck... Via had a good design by late 90s standards. Usable but pretty crap by the time they made the last ones in 2011, the dual core nano x2s. They are just not modern chips anymore, and they can't be made modern without legal issues.
 
DukenukemX said:
. Not to forget that Apple and Qualcomm are the only ones with fast ARM processors, and nobody else.
Click to expand...
Nvidia/Amazon certainly has fast arm processor, Ampere would be close has well, I imagine ampereOne M 192 cores-12 channel of ddr5 would quality.

Do not know about those Google axion, microsoft cobalt and so on, but I imagine they are not slow.

Has for PC processor, we could see some in road with Nvidia N1x (mediatek arm) and its successor and if their workstation desktop offer scale down one day (currently it is a 72 core grace cpu)
 
Ididar said:
Maybe I don't know the right people, but who the heck is doing an entire video editing project with no external monitors on battery? Is this a thing now? People sitting in coffee shops doing major tasks and expecting batteries to last that long? Most of the people I know who run their laptops on battery all day long are kids who are just too lazy to plug the thing in. They'll run it down to 5% while sitting next to a wall plug and never think to plug it in. I can see a very edge case for people doing heavy tasks on battery but most battery stuff is people waiting for a flight, sitting in a coffee shop catching up on some emails, or other such light tasks. Normal people (in my world) don't do major tasks while running around being mobile.
Click to expand...
Numerous creators have to work on videos “in the field,” where a wall outlet isn’t guaranteed; think certain movie/TV shoots, news producers, even social/YouTube stars who have to edit minutes after capturing footage, no matter where they happen to be.

It’s true that many plug in, but they shouldn’t have to for a short edit.
 
ChadD said:
I am fairly sure that is exactly what it actually is.
Its not as crazy as it sounds. If you don't already own a ARM license you do have to buy it. If you already own a Atom license which they do.... I mean why not. Zero costs vs cost is pretty easy math. For a company that doesn't already have that old license in their back pocket sure, ARM all the way.
Click to expand...
Developing an x86 CPU isn't cheap, especially the stuff I'm hearing in this new chip. So far it has 16 and 32 tile AMX, APX, FRED, x86S, and other engineering feats. It might be made for servers since it's X86S? It makes far more sense to pay for an ARM license. RISC-V is free and already somewhat works.

People on Phoronix might have already figured it out, and it maybe RosaicLabs. Intel gave them their blueprints for ATOM chips. I think we found our mystery engineers. They think Intel is thinking of selling their x86 license where I think Intel is hunting for new engineers. Makes sense since whoever is engineering Intel's chips have been sucking at it for at least 10 years. Could also be Intel is actually trying to make an x86 chip that's small and efficient enough for phones and tablets? As much of an improvement Panther Lake is for Intel, it's still a good generation or two behind AMD, and maybe 2-3 generations behind Apple. Doesn't seem like Intel fired enough of their bad engineers.
LukeTbk said:
Nvidia/Amazon certainly has fast arm processor, Ampere would be close has well, I imagine ampereOne M 192 cores-12 channel of ddr5 would quality.
Click to expand...
AMD has consistently beat everyone with their server chips. For desktop machines, these wouldn't be fast enough.
jbltecnicspro said:
As far as I'm concerned, ARM is the best. Cause it was in Gameboy Advance. Fuck all you haters.
Click to expand...
I wanna point out that the GameBoy Adance also includes a Z80 from Sharp SM83, which is based on the Intel 8080. It's there for backwards compatibility for the Gameboy. It's not x86 but it's a good two generations behind x86. There's no escape from x86.
 
Aurelius said:
Numerous creators have to work on videos “in the field,” where a wall outlet isn’t guaranteed; think certain movie/TV shoots, news producers, even social/YouTube stars who have to edit minutes after capturing footage, no matter where they happen to be.

It’s true that many plug in, but they shouldn’t have to for a short edit.
Click to expand...
Yeah that’s true but they are also likely travelling with a Bluetti or Jackery.
 
DukenukemX said:
Developing an x86 CPU isn't cheap, especially the stuff I'm hearing in this new chip. So far it has 16 and 32 tile AMX, APX, FRED, x86S, and other engineering feats. It might be made for servers since it's X86S? It makes far more sense to pay for an ARM license. RISC-V is free and already somewhat works.

People on Phoronix might have already figured it out, and it maybe RosaicLabs. Intel gave them their blueprints for ATOM chips. I think we found our mystery engineers. They think Intel is thinking of selling their x86 license where I think Intel is hunting for new engineers. Makes sense since whoever is engineering Intel's chips have been sucking at it for at least 10 years. Could also be Intel is actually trying to make an x86 chip that's small and efficient enough for phones and tablets? As much of an improvement Panther Lake is for Intel, it's still a good generation or two behind AMD, and maybe 2-3 generations behind Apple. Doesn't seem like Intel fired enough of their bad engineers.

AMD has consistently beat everyone with their server chips. For desktop machines, these wouldn't be fast enough.

I wanna point out that the GameBoy Adance also includes a Z80 from Sharp SM83, which is based on the Intel 8080. It's there for backwards compatibility for the Gameboy. It's not x86 but it's a good two generations behind x86. There's no escape from x86.
Click to expand...
ARMs greatest strength is how easy it is to get into them, if you can get your own license you can have at it, and if you don’t there are catalogs of options for your use case.
But ARMs default designs all have a common problem, being PCIe lane support.
Off the shelf you are maybe getting 4 lanes, 8 if you go top end. But the base Atom will get you 20.

Robots and such need a lot of inputs and base ARM options will leave you starved for IO, and custom options are expensive.

But ARM is gaining traction and adoption around the world, Intel needs to step up and licensing some small x86 options like the ATOMs is a step they should have taken a decade ago.
 
DukenukemX said:
Developing an x86 CPU isn't cheap, especially the stuff I'm hearing in this new chip. So far it has 16 and 32 tile AMX, APX, FRED, x86S, and other engineering feats. It might be made for servers since it's X86S? It makes far more sense to pay for an ARM license. RISC-V is free and already somewhat works.

People on Phoronix might have already figured it out, and it maybe RosaicLabs. Intel gave them their blueprints for ATOM chips. I think we found our mystery engineers. They think Intel is thinking of selling their x86 license where I think Intel is hunting for new engineers. Makes sense since whoever is engineering Intel's chips have been sucking at it for at least 10 years. Could also be Intel is actually trying to make an x86 chip that's small and efficient enough for phones and tablets? As much of an improvement Panther Lake is for Intel, it's still a good generation or two behind AMD, and maybe 2-3 generations behind Apple. Doesn't seem like Intel fired enough of their bad engineers.
Click to expand...

It is possible. I mean I just assumed x86S was dead as dead could be as Intel abandoned it with zero products a few years back already. Who knows though.
Thinking about it you know you might not be wrong.
This COULD possibly be a chip designed by a company like Microsoft or Amazon for their cloud servers. Perhaps this is Intel and AMD allowing one of the big boys who is tipping toward ARM domination some room to do something of their own and keep x86 relevant in cloud. (they could even license them specifically vendor locked... meaning no selling hardware to anyone else) Not saying it will work if that is the case. Thinking on it more though that could be a logical move. Amazon as an example is now adding 50% of new hardware as Graviton systems. Though most people suspect their x86 cloud is probably still 70% or so of their business. If things continue they are probably 1 or 2 graviton generations away from dropping x86 purchases down to next to zero anyway. Maybe letting them make their own in house x86 chip and getting at least a few fees in return is better then the alternative. Microsoft is essentially in the same position. 30% of their cloud sales are also ARM these days. Letting them make CPUs strictly for their own internal use as well might be a hail mary.

If Intel/AMD managed to get Amazon and MS to make CustomX86 chips rather then new ARM cloud servers. It could help stem ARM expansion beyond the big names. As neither Graviton nor Cobalt for being produced to be sold anyway. It seems a lot less insane for Intel/AMD to agree to license tech to be used in the same way. In house only. Instead of loosing a big chunk you gain some royalties at least, and keep x86 software development moving on the big platforms.

Regardless whatever is being cooked up it ain't for consumer use anyway. Though you could be right. MS or Amazon could 100% be the mystery maker. I could see GravitonX86 or CobaltX86 being a thing.
 
Last edited:
DukenukemX said:
I wanna point out that the GameBoy Adance also includes a Z80 from Sharp SM83, which is based on the Intel 8080. It's there for backwards compatibility for the Gameboy. It's not x86 but it's a good two generations behind x86. There's no escape from x86.
Click to expand...
Did you really take that post seriously? I didn't know about the Z80's relationship to 8080 though. :)
 
Aurelius said:
Numerous creators have to work on videos “in the field,” where a wall outlet isn’t guaranteed; think certain movie/TV shoots, news producers, even social/YouTube stars who have to edit minutes after capturing footage, no matter where they happen to be.

It’s true that many plug in, but they shouldn’t have to for a short edit.
Click to expand...

A short edit is not 10+ hours of video editing. If you are in a scenario where you have to do that much work in the field with no other choices then maybe you should have planned for a battery bank as well just in case. Also, movie/TV shoots and news producers have tons of power for every light/camera/etc they have on a shoot site so plugging in a laptop is hardly a risk. Also, they usually run around with monitors so people can more easily see what was shot instead of 5 people trying to huddle around a laptop screen.
 
  • Like
Reactions: kac77
like this
DukenukemX said:
AMD has consistently beat everyone with their server chips. For desktop machines, these wouldn't be fast enough.
Click to expand...
GB10 seem fast enough for a small desktop and it is a bit old cores by now, even compared with the big 16 cores AI Max+395 it does not look crazy:

https://www.phoronix.com/review/nvidia-gb10-cpu/6, it is a bit of desktop 13600k (both 20 thread chips) on Cinebench, when it has bad cooling in small box.

And Vera is about ~60% faster than grace, would they want on the N2X/Spark 2 gen to have fast enough for desktop CPU they certainly can, arguably can of already are.
 
ChadD said:
Thinking about it you know you might not be wrong.
Click to expand...
It is a mystery. It could be these chips end up in drones for all we know?
This COULD possibly be a chip designed by a company like Microsoft or Amazon for their cloud servers. Perhaps this is Intel and AMD allowing one of the big boys who is tipping toward ARM domination some room to do something of their own and keep x86 relevant in cloud. (they could even license them specifically vendor locked... meaning no selling hardware to anyone else) Not saying it will work if that is the case. Thinking on it more though that could be a logical move. Amazon as an example is now adding 50% of new hardware as Graviton systems. Though most people suspect their x86 cloud is probably still 70% or so of their business. If things continue they are probably 1 or 2 graviton generations away from dropping x86 purchases down to next to zero anyway. Maybe letting them make their own in house x86 chip and getting at least a few fees in return is better then the alternative. Microsoft is essentially in the same position. 30% of their cloud sales are also ARM these days. Letting them make CPUs strictly for their own internal use as well might be a hail mary.
Click to expand...
Licensing x86 has always been a good idea. I just don't know if AMD and Intel are ready to allow this? I'd like to see the market flooded with cheap x86 clones for SmartTV's and RaspberryPi's. This licensing move would be great during a time when people can't afford ram.
If Intel/AMD managed to get Amazon and MS to make CustomX86 chips rather then new ARM cloud servers. It could help stem ARM expansion beyond the big names. As neither Graviton nor Cobalt for being produced to be sold anyway. It seems a lot less insane for Intel/AMD to agree to license tech to be used in the same way. In house only. Instead of loosing a big chunk you gain some royalties at least, and keep x86 software development moving on the big platforms.
Click to expand...
Better that others can make x86 and lose that market dominance instead of losing it entirely to ARM.
Regardless whatever is being cooked up it ain't for consumer use anyway. Though you could be right. MS or Amazon could 100% be the mystery maker. I could see GravitonX86 or CobaltX86 being a thing.
Click to expand...
Even if it's meant for servers, it could still end up in consumers hands. It looks like it's meant to be efficient and small, which would work well for small devices. It also looks like it's made to have hundreds of CPU cores in a single die.
jbltecnicspro said:
Did you really take that post seriously? I didn't know about the Z80's relationship to 8080 though. :)
Click to expand...
No of course I knew you were being funny. I just like the history behind these chips. The Z80 which is based on the 8080 is also because it's made by the same man, Federico Faggin. A man that is so significant in computing history that I'm surprised he's not mentioned as much as Steve Jobs? It's interesting because during that time Japan was basically modern day China and they loved to copy chips. It's the reason why electronics were so cheap back then.
Federico Faggin intel 4004.jpg
 
Working at TI at the time, how can the US compete with Japan producing smaller for less? And thus began the beginning of the end of TI's silicon operations.
 
ChadD said:
It is possible. I mean I just assumed x86S was dead as dead could be as Intel abandoned it with zero products a few years back already. Who knows though.
Thinking about it you know you might not be wrong.
Click to expand...

Intel hanging onto Atom as long as they did makes sense in hindsight. Core was based on the Pentium M and it saved Intel's butt after AMD made its big push in the early 2000s. It's easy to overlook Atom because of how it was used at the time but it was damn impressive for what it was.
 
  • Like
Reactions: ChadD
like this
cjcox said:
Working at TI at the time, how can the US compete with Japan producing smaller for less? And thus began the beginning of the end of TI's silicon operations.
Click to expand...
It's why electronics were as cheap as they were during the 80's and 90's. It's probably why there was so much innovation because there wasn't anything limiting Japan.
Axman said:
Intel hanging onto Atom as long as they did makes sense in hindsight. Core was based on the Pentium M and it saved Intel's butt after AMD made its big push in the early 2000s. It's easy to overlook Atom because of how it was used at the time but it was damn impressive for what it was.
Click to expand...
Pentium M was nice but I think Athlon 64 was better. CoreDuo was also nice but was also only 32-bit. It wasn't until the Core2Duo when Intel had a product that could compete against AMD's Athlon II's. For a while AMD and Intel were nearly equal in performance until AMD released Bulldozer. AMD's Phenom II X6's were actually really nice chips. Sandybridge was what put AMD in their place until Ryzen. The last good Intel chip was Broadwell, and then everything Intel made afterwards just got hotter and more power hungry. Intel did make faster chips afterwards, but they just ate more power and got hotter in the process. The 9900K was a gamers choice CPU but the 2700X was just as fast and used 60 watts less power.

I'd imagine the reason for Atom was because it was a good baseline for something like x86S. As good as Panther lake is, it's also behind everyone else in performance and power. Here's hoping Nova Lake is actually good.
 
DukenukemX said:
I'd imagine the reason for Atom was because it was a good baseline for something like x86S.
Click to expand...
Not sure what we mean the reason for atom ? do we mean for people using it to make low power (robots, fridge, router, etc...) chip ?

I really doubt it is a good baseline for x86S (pure 64 bits, old bios method, old legacy instruction removed), older the cpu and more it would have been built with legacy instruction, old bios and 32 bits in mind ? Specially low power made for embedded device at the time, would have been more old school that desktop aiming cpu design. And x86s was it not been a bit cancelled ?

DukenukemX said:
It's why electronics were as cheap as they were during the 80's and 90's. It's probably why there was so much innovation because there wasn't anything limiting Japan.
Click to expand...
Not sure about the nothing limiting japan by the late 80s.... US government treats of tarrif and quota started mid 80s on Japanese electronics (https://en.wikipedia.org/wiki/1986_U.S.–Japan_Semiconductor_Agreement)

It felt very cheap to people that lived throught 1970s electronic pricing and because it was going down fast, but it was way more expensive than today, if you look at VCR, cd player/tv/anything today/phone vs calculator versus back then.... Nintendo NES got to be a bit cheaper than the XboxS once it was very mature and old by the late 80s, but for so much less.
 
DukenukemX said:
Sandybridge was what put AMD in their place until Ryzen. The last good Intel chip was Broadwell, and then everything Intel made afterwards just got hotter and more power hungry.
Click to expand...
I view most of the Core2Duo successors as security nightmares that made me seriously consider the the possibility that Intel cut a lot of corners to get the performance crown.
 
You must log in or register to reply here.
Back
Top