Thursday, August 14th 2025

Report: AMD Now Commands One-Third of the Desktop x86 Processor Market

Mercury Research compiled the latest data for the second quarter of 2025, with updates across the desktop, mobile, client, and server platforms. In Q2 2025, AMD achieved its best quarter yet. AMD's desktop x86 unit share rose to 32.2%, roughly one third of the market, and its desktop CPU revenue share climbed to 39.3%. Those gains reflect strong demand for Ryzen 9000 series processors and gamers' favorite X3D models, which together have driven both unit momentum and a disproportionate share of dollars in desktop sales. Note that Intel still remains the largest desktop vendor, but its unit share of 67.8% now represents a materially narrower lead than in prior years.

The desktop picture matters because value mix, not just units, is where margins are won. AMD is capturing more high-end sockets and achieving higher average selling prices, which helps its revenue share outpace its unit share. Intel continues to outsell AMD on desktops at a ratio of approximately two to one, down sharply from ratios that were far more present in earlier cycles. That compression signals a more contested market, with AMD converting enthusiast and commercial interest into concrete wins in both retail and OEM channels. Retail and DIY buyers are buying the latest Ryzen parts for gaming and content creation, while some system manufacturers are expanding Ryzen options in their higher-performance prebuilt lines.
Across the broader client landscape, AMD's total consumer unit share was 23.9% in Q2, up year-over-year, even as it slipped slightly quarter-over-quarter, and its consumer revenue share rose to 27.8%. Mobile unit share weakened for AMD relative to late 2024, but desktop strength and server gains are helping offset that softness. For Intel, the decline in desktop sales shows pressure at the higher end of its product stack as AMD narrows the gap through newer architectures and focused product placement. However, notebook share is an area where Intel is regaining some market share, making it an interesting trend to observe.
Here are the server unit shares too:
Source: Tom's Hardware

90 Comments on Report: AMD Now Commands One-Third of the Desktop x86 Processor Market

#76
Vayra86
john_I believe they lack manufacturing capacity to fill all those huge OEM orders. Also intel did gain the advantage with E cores, advertising higher number of cores. Probably intel Ultra 200 is also a much better option for laptops than for desktops where X3D CPUs are killing it. Still, even with taking in consideration all these, I was expecting AMD to keep gaining in the laptop market. But they need to get serious about it. Seeing Intel based Claw being on the news these last two days with optimizations making it look even faster than Z2 Extreme in same cases, I think at AMD they are taking their usual nap thinking that they can't lose from Intel in the iGPU game. And I hope they aren't doing the same with laptop SOCs.
Manufacturing capacity certainly plays a part.. great point. But not entirely - after all, why would AMD not be able to throw a huge bag of money at TSMC for first dibs on a new node or said capacity? Because- this is closely related to demand and represents a risk.
ARFBecause most people are either poor, or ill-informed about the benefits of better displays, or any combination of these reasons and others.
Of course that 1080p is the resolution of the poor and non-enthusiast | pro communities.

Hell, even if people work with larger drawings, be them AutoCAD or simple pdfs, everyone's needs are larger display with higher resolutions.
Or they just dont care ;) Not that many people really need more screen real estate and guess what, a large sum of those 4K panels are using 200% scaling in Windows..
Posted on Reply
#77
JustBenching
Vayra86Manufacturing capacity certainly plays a part.. great point. But not entirely - after all, why would AMD not be able to throw a huge bag of money at TSMC for first dibs on a new node or said capacity? Because- this is closely related to demand and represents a risk.
Because Apple amongst other companies can throw an even bigger bag I assume. Really don't know, how easy is it for TSMC to keep scaling their high end fabs? Don't they also face supply issues themselves?
Posted on Reply
#78
Vayra86
JustBenchingBecause Apple amongst other companies can throw an even bigger bag I assume. Really don't know, how easy is it for TSMC to keep scaling their high end fabs? Don't they also face supply issues themselves?
Why would their bag be bigger, I mean it is a guessing game but I dont see why Apple is special; the key difference between it and AMD is that Apple knows it sells those chips (maintain share and perhaps a tiny % of growth projected) and AMD would use it to gain market share it didnt have.
Posted on Reply
#79
JustBenching
Vayra86Why would their bag be bigger, I mean it is a guessing game but I dont see why Apple is special; the key difference between it and AMD is that Apple knows it sells those chips (maintain share and perhaps a tiny % of growth projected) and AMD would use it to gain market share it didnt have.
Cause apple is massive compared to AMD and I assume it has a much higher profit margin. Apple dwarfs Intel, AMD and nvidia combined in revenue if I remember correctly. It's freaking HUGE!
Posted on Reply
#80
Vayra86
JustBenchingCause apple is massive compared to AMD and I assume it has a much higher profit margin. Apple dwarfs Intel, AMD and nvidia combined in revenue if I remember correctly. It's freaking HUGE!
Youre right. 300 billion vs 50 ish or less per company ... wow
Posted on Reply
#81
Tek-Check
JustBenchingMy point is, if we agree that amd was slapping intel hard with zen 1 and zen 3 in MT performance, the fact that now intel not only has caught up but actually leading (by not a small margin) in a lot of segments clearly means that amd is not progressing, or in other words stagnating. This is not an opinion, these are just facts. Even in ST intel is ahead, but of course not with a sizable margin.
Interesting denial of reality. Just two days ago, Tom's Hardware retested the flagships for productivity. The verdict is clear. One just needs to comprehend in English to be able to digest it.
www.tomshardware.com/pc-components/cpus/amd-ryzen-9-9950x-vs-intel-core-ultra-9-285k-faceoff-it-isnt-even-close

Plus, this very website published findings for another flagship, 9950X3D. Flagships carry the message for its generation. Differences are even higher in Linux, seen in comprehensive review by Phoronix. It's not even close.
www.techpowerup.com/review/amd-ryzen-9-9950x3d/26.html
JustBenchingAs far as I understand the graphs represent sales. Units sold. And if that is the case then the majority of customers come to the conclusion that Intel systems are better for them. I don't like using sales as an indication of what product is good or not but since that's what you are doing, clearly intel has better products.
If you had commanded over 90% of market share until a few years and today you command around 67% (even less, 60% according to PassMark), the question is what has happened with all those "better" products and why people buy more of "worse" products from the competitor? If you still sell better products, why is it that you cannot keep market share intact? You are still selling majority of CPUs overall, but customers are less enthused by your recent generations, for whatever reason, hence your market share is slowly, but consistently melting. This process takes years, as you know. So, the narrative that Intel has better products just because they still sell more doesn't work, I am afraid.

Besides, one of key reasons why Intel still holds higher share in server CPU is due to long cycle of rack replacements (7-8 years) and historically high CPU attachment number. Majority of companies that decided to upgrade old servers replaced old Xeon systems with EPYCs, hence such rapid increase in EPYC revenues and installation base. Parity in share is expected by the end of 2026 or in 2027. Just watch it happen with Venice and Verano CPUs. We will come back to this debate again when next quarter numbers are released, and then nwxt year, and so on. You will continue to see a gradual decline, unless something big happens, such as Nova Lake and Diamond Rapids bringing winning flagships and revived enthusiasm across the board.
Posted on Reply
#82
JustBenching
Tek-CheckInteresting denial of reality. Just two days ago, Tom's Hardware retested the flagships for productivity. The verdict is clear. One just needs to comprehend in English to be able to digest it.
www.tomshardware.com/pc-components/cpus/amd-ryzen-9-9950x-vs-intel-core-ultra-9-285k-faceoff-it-isnt-even-close
.
www.techpowerup.com/review/amd-ryzen-9-9950x3d/26.html
It is clear indeed, not going to deny it. The CPUs were practically identical according to the reviewer who's review you just posted, one just needs to comprehend English.

Gaming : Similar 1% lows, 2% better averages for the amd, better efficiency for Intel.

ST : Intel flat out wins by 6.2%

MT : AMD leads by 2.3%!

Efficiency : AMD leads by 10-15% in MT scenarios, Intel leads by a whooping 60% in mixed scenarios.

Please, explain to me how that's a clear lead for AMD? Especially considering their starting position with zen 1, which was like 40% faster than the equivalent Intel part. Come on man, you are not being honest here.
Tek-CheckIf you had commanded over 90% of market share until a few years and today you command around 67% (even less, 60% according to PassMark), the question is what has happened with all those "better" products and why people buy more of "worse" products from the competitor? If you still sell better products, why is it that you cannot keep market share intact? You are still selling majority of CPUs overall, but customers are less enthused by your recent generations, for whatever reason, hence your market share is slowly, but consistently melting. This process takes years, as you know. So, the narrative that Intel has better products just because they still sell more doesn't work, I am afraid.
They had 90% marketshare when AMD wasn't making any CPUs my man. Obviously, once AMD started making CPUs again Intel wouldn't retain that 90%. How could they...
Posted on Reply
#83
Tek-Check
JustBenchingBecause Apple amongst other companies can throw an even bigger bag I assume. Really don't know, how easy is it for TSMC to keep scaling their high end fabs? Don't they also face supply issues themselves?
They do, of course, face gradual rampup for the most advanced node. It doesn't happen over night. There are several articles you can find to check N2 capacity increase as they build new fabs and production lines. Nvidia, Apple, AMD and a few others have already booked almost all N2 capacity for the next 18 months for their future architectures. Intel will also produce several tiles on TSMC as the confidence in A18 has not been stellar for several reasons.
Posted on Reply
#84
Dr. Dro
Tek-CheckIf you had commanded over 90% of market share until a few years and today you command around 67% (even less, 60% according to PassMark), the question is what has happened with all those "better" products and why people buy more of "worse" products from the competitor? If you still sell better products, why is it that you cannot keep market share intact? You are still selling majority of CPUs overall, but customers are less enthused by your recent generations, for whatever reason, hence your market share is slowly, but consistently melting. This process takes years, as you know. So, the narrative that Intel has better products just because they still sell more doesn't work, I am afraid.

Besides, one of key reasons why Intel still holds higher share in server CPU is due to long cycle of rack replacements (7-8 years) and historically high CPU attachment number. Majority of companies that decided to upgrade old servers replaced old Xeon systems with EPYCs, hence such rapid increase in EPYC revenues and installation base. Parity in share is expected by the end of 2026 or in 2027. Just watch it happen with Venice and Verano CPUs. We will come back to this debate again when next quarter numbers are released, and then nwxt year, and so on. You will continue to see a gradual decline, unless something big happens, such as Nova Lake and Diamond Rapids bringing winning flagships and revived enthusiasm across the board.
Yup, I agree. Without really going into the server realm (it's beyond my grasp and I'll admit I'm out of date there), Arrow Lake is, in my opinion, a failed product, especially for all that marketing fanfare and the rebranding of "Ultra" they came up with.

It brings some industry firsts like CKD RAM support, but there were way too many performance regressions, caveats and tradeoffs for it to truly be an upgrade over its predecessors. The only CPU I truly like is the U9 285 (non K), which can make an amazing, low maintenance machine for the right price, the problem being that the price hasn't been right for any of the U9 CPUs all this time, and most people overall will be better served by a 7800X3D/9800X3D with about the same power and thermal footprint, but much better gaming muscle.

As it stands I'd say Intel is a solid 5 years behind AMD in CPU technology right now, thankfully the very large, generational level performance advantage they had over Zen 3 at the time mostly covers up for this - a good 13th/14th Gen i9, and to that extent the U9 285/285K is not going to look too bad next to a 7950X, and the U7 265K more than holds its own at the pricing being practiced against the Ryzen 7 line. The 9950X/9950X3D are straight up faster overall, and I'm all too happy to concede that, caveats regarding the 9950X3D notwithstanding.
Posted on Reply
#85
JustBenching
Dr. DroYup, I agree. Without really going into the server realm (it's beyond my grasp and I'll admit I'm out of date there), Arrow Lake is, in my opinion, a failed product, especially for all that marketing fanfare and the rebranding of "Ultra" they came up with.

It brings some industry firsts like CKD RAM support, but there were way too many performance regressions, caveats and tradeoffs for it to truly be an upgrade over its predecessors. The only CPU I truly like is the U9 285 (non K), which can make an amazing, low maintenance machine for the right price, the problem being that the price hasn't been right for any of the U9 CPUs all this time, and most people overall will be better served by a 7800X3D/9800X3D with about the same power and thermal footprint, but much better gaming muscle.

As it stands I'd say Intel is a solid 5 years behind AMD in CPU technology right now, thankfully the very large, generational level performance advantage they had over Zen 3 at the time mostly covers up for this - a good 13th/14th Gen i9, and to that extent the U9 285/285K is not going to look too bad next to a 7950X, and the U7 265K more than holds its own at the pricing being practiced against the Ryzen 7 line. The 9950X/9950X3D are straight up faster overall, and I'm all too happy to concede that, caveats regarding the 9950X3D notwithstanding.
My observation is - AMD started from a 199$ CPU beating i5's and i7's in MT performance to today, where their 280$ 9600x loses in both MT and overall application performance to 4 year old i7s and every i5 in existence. I don't get how this is progress compared to where they started. The R5 1600 was close to the 7700k in overall performance, nowadays it's nowhere near the 265k. And that's considered a failed CPU generation from Intel, lol.. We went from an i7 being only 13% faster than the 199$ MSRP R5 1600 to an i7 being 50% faster than a 280$ MSRP R5 9600x.




Posted on Reply
#86
Dr. Dro
JustBenchingMy observation is - AMD started from a 199$ CPU beating i5's and i7's in MT performance to today, where their 280$ 9600x losing in both MT and overall application performance to 4 year old i7s and every i5 in existence. I don't get how this is progress compared to where they started. The R5 1600 was close to the 7700k in overall performance, nowadays it's nowhere near to the 265k. And that's considered a failed CPU generation from Intel, lol.




Yes, it's true, AMD has been price creeping as they gain mind and market, and they haven't shied away from charging whatever they want on HEDT - Threadripper's a good glimpse of what the CPU market will look like without Intel in the picture, $11,700 flagships and all.

It's no surprise the 265K can often pound the 9600X and most other 8 core R7s with its 8P+12E setup, but with all the negative press and the promise of yet another generation on AMD's side, a lot of people will opt for the 9600X on that alone. The thing about Intel vs. AMD is that each type of architecture still holds an advantage at something, for example, even that 9600X is gonna look as good if not better than the 265K if your workload uses AVX-512, so it's always good to have choice, especially since to have it all right now, you're going to have to go with Threadripper.

Intel still holds the tweakers' advantage, you can extract a lot more out of the Intel chips than you can out of the AMD ones even after the release of the X870E Apex board (mostly because only their APU is tweakable to that extent), but it's only a matter of time until AMD is no longer bound by the IOD and offers the same level of tweaking ability as Intel.

I just wonder why are the U9s so bloody expensive when Intel seems able to keep relatively good stock on the 265K, for example the U9 285 costs almost double here and is almost always out of stock, the 285K then, a complete unicorn, I think I've saw 2 shipments of it since it launched, that doesn't make sense to me. Intel needed to flood the market with cheap U9 chips, but being TSMC N3B, I guess they just can't.
Posted on Reply
#87
Tek-Check
JustBenchingThey had 90% marketshare when AMD wasn't making any CPUs my man. Obviously, once AMD started making CPUs again Intel wouldn't retain that 90%. How could they...
No. AMD didn't manufacture good enough products for several years before Zen to make people feel enhusiastic about their CPUs and buy them together with platforms those CPUs worked in. I know this because I bought one of Sandy Bridge systems with legendary i7 2700K and called it a day for several years.

Fast forward, the longevity of both AM4 and AM5 platforms has proven one of key winning strategies in reclaiming market share, alongside improvements in CPUs. To be clear, not all CPUs working in those platforms are as good in performance and value, but there are enough releases that allow a lot of choices for consumers to mix-and-match. X3D CPUs have sped up the loss of market share by Intel and that's another key factor.

AMD was able to offer three lines of desktop CPUs (G APU, vanilla and X3D) while Intel continued to offer one generic line of CPUs that do it all. This diverse offer helped customers buy more specialised CPU for their daily needs, be it gaming, productivity, graphics or mix of any of those. Plus upgrade cycles.

It's not surprising that we are already hearing rumours that Nova Lake will bring proper multi-generational support and that even some Zen7 SKUs might release on AM5 platform, to further boost the longevity of the platform. We will see how this pans out.
Posted on Reply
#88
JustBenching
Dr. DroYes, it's true, AMD has been price creeping as they gain mind and market, and they haven't shied away from charging whatever they want on HEDT - Threadripper's a good glimpse of what the CPU market will look like without Intel in the picture, $11,700 flagships and all.

It's no surprise the 265K can often pound the 9600X and most other 8 core R7s with its 8P+12E setup, but with all the negative press and the promise of yet another generation on AMD's side, a lot of people will opt for the 9600X on that alone. The thing about Intel vs. AMD is that each type of architecture still holds an advantage at something, for example, even that 9600X is gonna look as good if not better than the 265K if your workload uses AVX-512, so it's always good to have choice, especially since to have it all right now, you're going to have to go with Threadripper.

Intel still holds the tweakers' advantage, you can extract a lot more out of the Intel chips than you can out of the AMD ones even after the release of the X870E Apex board (mostly because only their APU is tweakable to that extent), but it's only a matter of time until AMD is no longer bound by the IOD and offers the same level of tweaking ability as Intel.

I just wonder why are the U9s so bloody expensive when Intel seems able to keep relatively good stock on the 265K, for example the U9 285 costs almost double here and is almost always out of stock, the 285K then, a complete unicorn, I think I've saw 2 shipments of it since it launched, that doesn't make sense to me. Intel needed to flood the market with cheap U9 chips, but being TSMC N3B, I guess they just can't.
But that's the issue with core count deficiencies, AVX support sounds great on paper but then you realize that Intel offers so many more cores that they actually end up faster even in AVX accelerated workloads. IE stockfish makes excessive use of AVX, 265k is 50% faster than the 9700x.. It's the same crap in reverse, amd didn't have AVX but their 8 core 1700x ended up pounding the 7700k in AVX workloads due to sheer core count advantage. They badly need the 12c CCD and I hope the rumors about it are true.
Posted on Reply
#89
Tek-Check
JustBenchingCause apple is massive compared to AMD and I assume it has a much higher profit margin. Apple dwarfs Intel, AMD and nvidia combined in revenue if I remember correctly. It's freaking HUGE!
Apple has also lost whopping 1 trillion dollar in valuation recently, so their situation is in flux right now, especially after being sued for misadvertising non-existent AI integration.
Posted on Reply
#90
the54thvoid
Super Moderator
So far off topic. Locked. There are other forums you can swing your willies about in.
Posted on Reply
Nov 13th, 2026 23:20 CST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts