AMD: We Stand Ready To Make Arm Chips

[Lightwreather]

Summary

 AMD's CFO Devinder Kumar recently commented that AMD stands ready to manufacture Arm chips if needed, noting that the company's customers want to work with AMD on Arm-based solutions. 

Quotes

Quote

Kumar's remarks came during last week's Deutsche Bank Technology Conference, but they built on comments from AMD CEO Lisa Su earlier in the year that underscored the company's willingness to create custom silicon solutions for its customers, be they with x86 or Arm cores.
"But I'll tell you from my standpoint, when you look at compute solutions, whether it's x86 or ARM or even other areas, that is an area for our focus on investment for us," Kumar responded to a question about the company's view of competing Arm chips. "We know compute really well. Even ARM, as you referenced, we have a very good relationship with ARM. And we understand that our customers want to work with us with that particular product to deliver the solutions. We stand ready to go ahead and do that even though it's not x86, although we believe x86 is a dominant strength in that area."
Kumar's comments don't give us any concrete indication that the company has specific Arm projects underway. Still, they do confirm that the company's customers are interested in purchasing Arm-based processors from AMD, and that the company is willing to fulfill that need.
AMD could deliver Arm chips through several possible go-to-market avenues, such as through standard off-the-shelf data center or desktop PC processor lineups as an AMD-branded solution, or through its semi-custom business, wherein it would design Arm processors to meet certain customer criteria. Given the trend we've seen of an increasing number of large hyperscalers developing their own chips, like AWS with Graviton, it's reasonable to expect that AMD could attract plenty of interest for Arm designs through its semi-custom business.
AMD CEO Lisa Su set the tone earlier this year at the JPMorgan Global Technology event, saying, "I think AMD has a lot of experience with the ARM architecture. We have done quite a bit of design in our history with ARM as well. We actually consider ARM as a partner in many respects."

"From an AMD standpoint, we consider ourselves sort of the high-performance computing solution working with our customers, and that that is certainly the way we look at this. And if it means ARM for certain customers, we would certainly consider something in that realm as well," Su explained. 

AMD is an Arm licensee and has quite a bit of experience with Arm architectures,  dating back to its K12 architecture that never came to market as planned back in 2017. Fast forward to today, and AMD does ship Arm cores, but they come as small microcontrollers for relatively simple tasks, like the company's in-built Platform Security Processors (PSP) that perform security functions to harden the company's CPUs.
AMD is willing to diversify to remain competitive in all possible workloads, and it has an expansive IP canvas to paint on. For instance, AMD could use its Infinity Fabric to tie numerous elements, such as chiplets, GPU cores, and various memories, into one cohesive unit. Additionally, its pending Xilinx acquisition could provide plenty of avenues for FPGA additives.
It's easy to imagine AMD creating powerful data center-class chips for its clients, but even though Arm has reached record market share in PCs, it doesn't seem likely that we'll see AMD add Arm-powered PC chips to its stable any time soon.
Naturally, the fog of Nvidia's pending $45 billion Arm acquisition hangs thick over any conversation about Arm, but that endeavor has hit more than a few snags and might not pass regulatory muster. AMD and Nvidia haven't always been the best of chums, but Nvidia has pledged to honor Arm's commitments and to continue licensing the architecture, even to its rivals, if the deal does go through. That could shape up for an odd encounter in the data center, with AMD-designed Arm chips facing off with Nvidia's Arm-based Grace CPUs.
It could get even stranger, though. Intel, the other half of what is effectively an x86 duopoly, has pledged to produce chips based on just about any architecture through its Intel Foundry Services, be they Arm or based on the open RISC-V architecture. Intel may take things a step further, though, with its rumored $2 billion acquisition attempt of RISC-V chipmaker SiFive, proving that the coming years could completely redefine the semiconductor market.
Meanwhile, AMD stands ready to punch out Arm-based chips for its customers, so it is probably more of a question of when it will happen rather than if.

 

My thoughts

Honestly, in my opinion this is great. This might push more competition in not only the Chips themselves, but the ISA as a whole, They would need both x86(x64) and ARM (aarch64) to be both compelling, so more inoovation. This might also lead to a standardisation of ARM, the fragmentation of which is oneof the gripes I have with ARM and RISC-V. Anyway, it should be nothed that intel also owns an ARM licenseand thus CAN manufacture their own arm chips, but I doubt they would. On the other hand with IDM 2.0, they are willing to do manufacture for others. So who knows, right now is a very exciting time in the tech space.

Sources

Tom's hardware

[Arika]

If only MS would get off their ass and put more effort into Windows on ARM.

 

Not sure what Linux is like on ARM, but given how fragmented the distros are already i can't imagine it would be a pleasant experience unless you had a very specific arm supported purpose

[Lightwreather]

1 minute ago, Arika S said:

Not sure what Linux is like on ARM, but given how fragmented the distros are already i can't imagine it would be a pleasant experience unless you had a very specific arm supported purpose

On the rpi and some other ARM devices (eg. Pinebook, etc.), Linux is pretty nice, you might have some trouble finding packages, but overall it's a pretty good experiece. There are even multiple distros, even some major ones, like ARch, Ubuntu, FEdora, etc.

[Eigenvektor]

15 minutes ago, Arika S said:

Not sure what Linux is like on ARM, but given how fragmented the distros are already i can't imagine it would be a pleasant experience unless you had a very specific arm supported purpose

Quite the opposite. Since the majority of Linux software is open source, it can simply be recompiled for ARM. Unlike Windows you don't have to wait for software vendors to make their apps available for ARM. So you not only have an OS, you actually have software that runs on it.

 

There's a number of distributions that support ARM already (as mentioned above). Their repositories for ARM are smaller than for x86 (for now) due to a lack of interest in the platform, but the most commonly used stuff should be there. And if you need something not yet available in pre-compiled form, you could theoretically compile it yourself if so inclined (provided you haven't found the one edge case where developers included hand crafted x86 assembly).

[porina]

1 hour ago, J-from-Nucleon said:

This might push more competition in not only the Chips themselves, but the ISA as a whole,

It is interesting, but usage of ARM based chips is quite different from usage of x86 chips, so the equally big question is what AMD intends to do with ARM? Also ARM is less about general purpose compute, and its usage is more into targeted use cases. Could AMD design a high(er) performance ARM core and re-license it out?

 

This is likely also a safety net for the future. x86 might not remain widely used indefinitely. Intel have a far more diverse product portfolio than AMD so would be less impacted by that.

 

1 hour ago, J-from-Nucleon said:

This might also lead to a standardisation of ARM, the fragmentation of which is oneof the gripes I have with ARM and RISC-V.

The way they're integrated into products as a whole makes the customisation of cores a strength in its own way. If you need a bit more of one thing, a bit less of another, you can do that. x86 you have a far more limited choice beyond picking what you're offered. With AMD there isn't even a choice in core, only scaling the number of cores.

 

1 hour ago, J-from-Nucleon said:

Anyway, it should be nothed that intel also owns an ARM licenseand thus CAN manufacture their own arm chips, but I doubt they would. On the other hand with IDM 2.0, they are willing to do manufacture for others.

Intel are also looking closely at RISC-V, and I think if they're going to integrate non-x86 cores in their own future products, that'll probably be the way to go.

 

Don't have a link on hand, but check out the Anandtech interview with Jim Keller on CPU design. One of the more interesting comments within was a statement that the perceived advantages of ARM over x86, and even RSIC-V over ARM, is due to their age. x86 being the most mature of the lot has picked up more stuff over the years. ARM is well on its way there. The newcomer RISC-V is relatively young and pure, for now. It too will follow that path, and something else in future will be the next big thing.

[Lightwreather]

42 minutes ago, porina said:

Don't have a link on hand, but check out the Anandtech interview with Jim Keller on CPU design. One of the more interesting comments within was a statement that the perceived advantages of ARM over x86, and even RSIC-V over ARM, is due to their age. x86 being the most mature of the lot has picked up more stuff over the years. ARM is well on its way there. The newcomer RISC-V is relatively young and pure, for now. It too will follow that path, and something else in future will be the next big thing.

Sure, will do thanks.

[Doobeedoo]

Well that's neat, future will be very interesting in CPU and GPU space.

[Distinctly Average]

5 hours ago, porina said:

It is interesting, but usage of ARM based chips is quite different from usage of x86 chips, so the equally big question is what AMD intends to do with ARM? Also ARM is less about general purpose compute, and its usage is more into targeted use cases. Could AMD design a high(er) performance ARM core and re-license it out?

 

 

There are a really wide range of ARM uses. Plenty of specialised ones sure, but also a lot of general compute. Also remember that Apple M1 is ARM based.

[cj09beira]

5 hours ago, porina said:

It is interesting, but usage of ARM based chips is quite different from usage of x86 chips, so the equally big question is what AMD intends to do with ARM? Also ARM is less about general purpose compute, and its usage is more into targeted use cases. Could AMD design a high(er) performance ARM core and re-license it out?

 

This is likely also a safety net for the future. x86 might not remain widely used indefinitely. Intel have a far more diverse product portfolio than AMD so would be less impacted by that.

 

The way they're integrated into products as a whole makes the customisation of cores a strength in its own way. If you need a bit more of one thing, a bit less of another, you can do that. x86 you have a far more limited choice beyond picking what you're offered. With AMD there isn't even a choice in core, only scaling the number of cores.

 

Intel are also looking closely at RISC-V, and I think if they're going to integrate non-x86 cores in their own future products, that'll probably be the way to go.

 

Don't have a link on hand, but check out the Anandtech interview with Jim Keller on CPU design. One of the more interesting comments within was a statement that the perceived advantages of ARM over x86, and even RSIC-V over ARM, is due to their age. x86 being the most mature of the lot has picked up more stuff over the years. ARM is well on its way there. The newcomer RISC-V is relatively young and pure, for now. It too will follow that path, and something else in future will be the next big thing.

he also said it had little performance gains from its simpler ISA, as today's chips are so large a more complex decoder is trivially cheap to implement

[mr moose]

7 hours ago, Arika S said:

If only MS would get off their ass and put more effort into Windows on ARM.

 

Not sure what Linux is like on ARM, but given how fragmented the distros are already i can't imagine it would be a pleasant experience unless you had a very specific arm supported purpose

I've been using Linux on and off all the time Since 2003/4 (or thereabouts),  I currently have 2 systems with it and its generally my go to for PCs that are too good to toss but I don't want to buy windows for (I like to keep all my software legit).  And I have to say, for all the glory talk it gets I still encounter too many problems to recommend to anyone as a daily driver.   What works here doesn't work there,  work arounds for unsupported hardware can be a nightmare and random crashes with no error codes.    Unless something seriously changes in the linux camp I don't think we will ever see it go farther than it currently is (even with the dedicated distros for steam it's barely increasing market share). 

[igormp]

7 hours ago, Arika S said:

Not sure what Linux is like on ARM, but given how fragmented the distros are already i can't imagine it would be a pleasant experience unless you had a very specific arm supported purpose

I have both a laptop with arm and linux (a chromebook, actually), an arm cloud instance (graviton), a hackish phone with linux that I run some silly stuff on, and many other ARM SBCs, the only problem I face is when I need some proprietary stuff that has been only compiled for x86, other than that it works perfectly fine for me, but then again, none of those are used as a desktop, nor I'd consider myself a "regular" desktop user by this forum's standards.

[Fasterthannothing]

AMD: We will make Arm chips

Nvidia: Buys Arm - tells AMD they now work for them 

 

 

[Vishera]

I recommend Android for ARM desktops...

[Shreyas1]

3 hours ago, Vishera said:

I recommend Android for ARM desktops...

That's what I was thinking too. Is it possible to make a version of Android that's more tailored towards desktops rather than mobile devices?

[Curufinwe_wins]

This is a really odd position unless they simply don't have their own little cores to deal with the incoming transition to physically heterogenous core structures coming to desktop.

 

Ala having stopped Jaguar and further efforts. Also they *don't* manufacture ANY chips, which makes that wording really bizarre. Who knows, maybe Intel might actually be on to something with Alder Lake afterall.

[igormp]

3 minutes ago, Curufinwe_wins said:

This is a really odd position unless they simply don't have their own little cores to deal with the incoming transition to physically heterogenous core structures coming to desktop.

 

Ala having stopped Jaguar and further efforts. Also they *don't* manufacture ANY chips, which makes that wording really bizarre. Who knows, maybe Intel might actually be on to something with Alder Lake afterall.

At the time of Zen's conception, AMD had an modern ARM project (K12, as mentioned in the OP) along with a hybrid ARM/x86 project (skybridge), both led by Jim Keller, but then Zen had a tremendous success (enogh to warrant total focus on it) and Jim Keller left the company, but I guess they could simply continue from where they left off.

 

AFAIK, AMD has no other x86 µArch design available, even their low end Athlons and entry Ryzen CPUs were Zen-based.

 

I'm not sure if skybridge is meant as a single platform for multiple ISAs (as in, same motherboard/chipset for both ARM or x86 CPUs), or an actual heterogeneous CPU with both ARM and x86 cores on the same die (or even a single core with 2 front-ends/decoders sharing the execution units!!).

It's probably the former, but the latter sounds cool as heck, even if it's kinda unfeasible 

[LAwLz]

1 hour ago, Shreyas1 said:

That's what I was thinking too. Is it possible to make a version of Android that's more tailored towards desktops rather than mobile devices?

It's possible, but that's kind of what ChromeOS is for. It runs android apps too if that's what you're after. 

[Curufinwe_wins]

3 hours ago, igormp said:

At the time of Zen's conception, AMD had an modern ARM project (K12, as mentioned in the OP) along with a hybrid ARM/x86 project (skybridge), both led by Jim Keller, but then Zen had a tremendous success (enogh to warrant total focus on it) and Jim Keller left the company, but I guess they could simply continue from where they left off.

 

AFAIK, AMD has no other x86 µArch design available, even their low end Athlons and entry Ryzen CPUs were Zen-based.

 

I'm not sure if skybridge is meant as a single platform for multiple ISAs (as in, same motherboard/chipset for both ARM or x86 CPUs), or an actual heterogeneous CPU with both ARM and x86 cores on the same die (or even a single core with 2 front-ends/decoders sharing the execution units!!).

It's probably the former, but the latter sounds cool as heck, even if it's kinda unfeasible 

Fair enough, yeah I don't think they did anything after Puma with the Jaguar line, so that should be absolutely worthless in a modern era.

[Donut417]

17 hours ago, Arika S said:

 

If only MS would get off their ass and put more effort into Windows on ARM.

 

I blame Qualcomm more than I blame Microsoft. If Microsoft had a better option they would probably had a better experience. Qualcomm has about 0 stakes in the desktop market, so do you think they give a shit about making the best desktop chip? No, because they pretty much own a good chunk of the mobile market. AMD on the other had has a mighty big stake in the desktop CPU market. So this could very well be the saving grace for Windows on ARM. Because I could honestly see AMD really pushing ARM to the limits if they could. 

 

The biggest issue with Windows on ARM is legacy support. Thats why even if ARM is successful in the PC market, PC's are likely still to be dominated by x86 for the foreseeable future. 

[igormp]

1 hour ago, Donut417 said:

I blame Qualcomm more than I blame Microsoft. If Microsoft had a better option they would probably had a better experience. Qualcomm has about 0 stakes in the desktop market, so do you think they give a shit about making the best desktop chip? No, because they pretty much own a good chunk of the mobile market. AMD on the other had has a mighty big stake in the desktop CPU market. So this could very well be the saving grace for Windows on ARM. Because I could honestly see AMD really pushing ARM to the limits if they could. 

 

The biggest issue with Windows on ARM is legacy support. Thats why even if ARM is successful in the PC market, PC's are likely still to be dominated by x86 for the foreseeable future. 

Samsung could also be a player in this scenario, since their exynos CPUs are okaish, and they already do laptops (even some ARM-based ones).

[LAwLz]

8 hours ago, Donut417 said:

I blame Qualcomm more than I blame Microsoft. If Microsoft had a better option they would probably had a better experience. Qualcomm has about 0 stakes in the desktop market, so do you think they give a shit about making the best desktop chip? No, because they pretty much own a good chunk of the mobile market. AMD on the other had has a mighty big stake in the desktop CPU market. So this could very well be the saving grace for Windows on ARM. Because I could honestly see AMD really pushing ARM to the limits if they could. 

 

The biggest issue with Windows on ARM is legacy support. Thats why even if ARM is successful in the PC market, PC's are likely still to be dominated by x86 for the foreseeable future. 

I think it's about 50/50 who to blame.

 

Do I blame Qualcomm? Absolutely. The SoCs have just not been that good. Compared to Apple they are constantly 2-3 generations behind, if not more. Hopefully that will change with their recent acquisition of Nuvia, but even if it does, we're still a couple of years away from end products in the hands of consumers.

 

But here is why I think Microsoft shares an equal amount of blame.

1) Microsoft has encouraged an ecosystem of lazy and closed source developers. As a result, existing software is really hard to port to ARM, and not many developers care to do it. Look at GNU/Linux for example, it has a way bigger portfolio of ARM software than Windows has.

 

2) Window on ARM's x86 emulation is extremely slow compared to Apple's. Some people, including LTT, have said that this is because Apple has special hardware in their SoC. This is kinda true but not in the way people think. The instructions Apple uses to speed up x86 emulation are just standard ARM instructions. It's just that neither Microsoft not Qualcomm been bothered to implement them. Window on ARM's x86 emulator was written for the Snapdragon 820, and it does not seem like it has been updated that much since. So the emulator targets the Armv8.0-A ISA. The instructions Apple uses to make their emulator so fast were introduced in Armv8.3-A.

Microsoft really needs to rewrite their emulator with the newer and faster instructions and work together with for example Qualcomm to make processors that supports it. 

When it comes to intricate hardware emulation, you can't just write something for a 6 year old CPU and then go "well, that's probably good enough". Hell, we still don't have 64bit emulation. That's coming with Windows 11. That should have been available on day 1.

 

3) Microsoft has not been serious about Windows on ARM. They have not pushed it, or pushed it well. They quickly hacked it together and went "look it works" and then didn't really bother with it for ages.

 

4) Windows is a horrible mess of spaghetti code which makes it near impossible to work on. That's why porting it to ARM has been such a challenge for Microsoft. Meanwhile, both MacOS and GNU/Linux had no problems at all being ported. Sure, the developers poured a lot of effort into it, but in both cases it seems to have been far smoother than with Windows.

 

5) Microsoft are really lazy when it comes to development, and not even their own programs are well written. If they spent some time and effort into making at least their own programs work decently on x86 and ARM, it wouldn't be as painful to use on slower devices. As it stands, even their chat program (Teams) and text editor (VSCode) are honestly some of the worst written pieces of software I have ever had to use. It is seriously laughable how poorly written they are. A while ago Microsoft announced that by swapping out Electron for WebView2, they cut memory usage in half... But it's still way higher than what it should be. It is seriously embarrassing that a company as large as Microsoft has to resort to really poor performing Electron/WV2 and React based software. I download and install software because they are suppose to offer higher performance than web apps. I don't download and install a program to get a shitty web app.

 

 

So to summarize.

Microsoft has in the past encouraged developers to make their programs hard to port. This is now coming back and biting them in the ass when programs needs to be ported. Microsoft don't support hardware features that would make their emulator faster. Microsoft fucked up the Windows on ARM launch by doing their terrible "let's ship in in a broken state and then fix it afterwards". This causes consumers to taste it once, think it's terrible, and then not try it again. Windows is terribly written. Microsoft's own first party programs are very resource intense because Microsoft are lazy and as a result a lot of programs that SHOULD work well on low-end hardware (like the ARM processors Windows on ARM uses) don't work well.

[leadeater]

18 hours ago, igormp said:

At the time of Zen's conception, AMD had an modern ARM project (K12, as mentioned in the OP) along with a hybrid ARM/x86 project (skybridge), both led by Jim Keller, but then Zen had a tremendous success (enogh to warrant total focus on it) and Jim Keller left the company, but I guess they could simply continue from where they left off.

 

AFAIK, AMD has no other x86 µArch design available, even their low end Athlons and entry Ryzen CPUs were Zen-based.

 

I'm not sure if skybridge is meant as a single platform for multiple ISAs (as in, same motherboard/chipset for both ARM or x86 CPUs), or an actual heterogeneous CPU with both ARM and x86 cores on the same die (or even a single core with 2 front-ends/decoders sharing the execution units!!).

It's probably the former, but the latter sounds cool as heck, even if it's kinda unfeasible 

AMD also had some past efforts with ARM too. Problem is nothing has ever really come of either of their bigger two attempts at it so I don't really know how good their ARM products are or could have been.

 

I'm sure it wouldn't be bad because anyone looking in to ARM SoC development has access to good IP and designs anyway and AMD's teams are good at what they do.

 

I have a feeling AMD is purposely playing a background role when it comes to ARM because I think their market assessment is that they have better business prospects within the x86 ecosystem due to the far more closed nature of it. If AMD started pushing more ARM then that also invites more companies to compete in the market with, where as x86 is just AMD Intel and VIA.

[igormp]

5 hours ago, LAwLz said:

The instructions Apple uses to speed up x86 emulation are just standard ARM instructions. It's just that neither Microsoft not Qualcomm been bothered to implement them.

It's just just "regular arm instructions", those are only used to toggle the memory model. The TSO emulation that Apple implemented, and which can be toggled on or off during runtime, is what makes rosetta so fast.

It could be implemented, sure, but then it would require some major design changes to cache coherency, buffers and overall memory management inside a core and goes beyond the ISA itself.

5 hours ago, LAwLz said:

text editor (VSCode) are honestly some of the worst written pieces of software I have ever had to use. It is seriously laughable how poorly written they are. A while ago Microsoft announced that by swapping out Electron for WebView2, they cut memory usage in half... But it's still way higher than what it should be. It is seriously embarrassing that a company as large as Microsoft has to resort to really poor performing Electron/WV2 and React based software.

VSCode is one of the most (if not THE most) performant electron programs out there, doesn't change the fact that electron is shit, but it's really easy to extend and that's why it has so many users, the myriad of plugins is something that you won't find anywhere else.

I still miss sublime, but now I have more than enough ram to use VSCode with all of the nice plugins available.

5 hours ago, LAwLz said:

3) Microsoft has not been serious about Windows on ARM.

It bummers me how MS used to have many ports to weird archs, be it PowerPC, IA-64, Alpha or even MIPS of all things, and now they can't even port their system to the most famous ISA used in mobiles.

 

 

[LAwLz]

2 hours ago, igormp said:

It's just just "regular arm instructions", those are only used to toggle the memory model. The TSO emulation that Apple implemented, and which can be toggled on or off during runtime, is what makes rosetta so fast.

It could be implemented, sure, but then it would require some major design changes to cache coherency, buffers and overall memory management inside a core and goes beyond the ISA itself.

But it is regular ARM instructions.

What exactly do you think TSO is? It changes the memory model to be similar to x86. What do the instructions I referred to do? Makes the memory model similar to x86. It's those instructions that makes Rosetta so fast. Apple is one of the few hardware makers that have implemented them and as far as I know, the only consumer software maker that uses them.

TSO is just Apple's name for the OS functions that uses those instructions.

 

Sure it would require some hardware changes but that's like saying "AVX requires hardware changes". Of course supporting another instruction requires hardware changes. My point is that Rosetta is fast because of standard Arm instructions. It's not black magic. They are just using the tools that are available to everyone.

 

 

2 hours ago, igormp said:

VSCode is one of the most (if not THE most) performant electron programs out there, doesn't change the fact that electron is shit, but it's really easy to extend and that's why it has so many users, the myriad of plugins is something that you won't find anywhere else.

I still miss sublime, but now I have more than enough ram to use VSCode with all of the nice plugins available.

Yeah... So it's the best out of a bunch of shit. Not exactly a strong endorsement there.

Sublime is so much faster and lighter weight it's crazy. VSCode could be that fast if Microsoft gave a crap. But their current policy seems to be "make it just good enough".

 

 

2 hours ago, igormp said:

It bummers me how MS used to have many ports to weird archs, be it PowerPC, IA-64, Alpha or even MIPS of all things, and now they can't even port their system to the most famous ISA used in mobiles.

Well, that was like 30 years ago when NT was still new. If you mistreat your codebase for like 30 years by writing poor code, poor documentation and always settling for "good enough" then it becomes an unwieldy and inflexible monster, which is exactly what has happened to Windows.

[igormp]

3 hours ago, LAwLz said:

But it is regular ARM instructions.

What exactly do you think TSO is? It changes the memory model to be similar to x86.

AFAIK, it's related to how you deal with memory/cache coherency between registers/cores/caches/memory.

3 hours ago, LAwLz said:

What do the instructions I referred to do?

Tbf, you didn't specify any instructions 

Anyway, TIL and found those on the ARM docs: https://developer.arm.com/documentation/den0024/a/Memory-Ordering/Memory-types

Apparently the equivalent to x86's TSO is called "Device-nGnRnE"

 

3 hours ago, LAwLz said:

Sublime is so much faster and lighter weight it's crazy

Yeah, but it lacks the whole ecosystem whilst being closed source and kinda expensive. For something fast that can manage multi-GB files with ease and has a nice ecosystem I justo resort to vim nowadays (I do still use VSCode 90% of the time).