Thursday, March 26th 2026

AMD Announces Ryzen 9 9950X3D2 Dual Edition Desktop Processor with 192 MB L3 Cache

AMD today launched the Ryzen 9 9950X3D2 Dual Edition Socket AM5 desktop processor. This processor is positioned above the Ryzen 9 9950X3D, and is the much awaited implementation of 3D V-Cache memory on both CCDs (CPU complex dies) of the processor. The 16-core/32-thread processor is based on the Zen 5 microarchitecture, and features a total of 192 MB of shared L3 cache—that's 32 MB of on-die L3 cache plus 64 MB of 3D V-Cache that's contiguous with it, across both CCDs. This enables the maximum possible gaming performance, as worker threads of the game can be located on either of the two CCDs and benefit from large last-level cache. The base clock is 4.30 GHz, with a boost clock of 5.60 GHz—100 MHz lower boost than the 9950X3D, and the same clocks as the 9850X3D—no word if these are the same for each CCD.

The AMD Ryzen 9 9950X3D2 Dual Edition is positioned to be a 16-core Zen 5 Flagship part, and to ensure the chip has not just the most cache, but also the highest possible clock speeds in the series, and retain its overclocking capabilities, AMD expanded the TDP of the processor to 200 W—the highest ever for an AMD Ryzen part. AMD is offering gaming performance rivaling its latest Ryzen 7 9850X3D part, but with significantly improved productivity performance in applications with large cached data-sets, such as software development, AI training and inferencing, and video production. The company's first-party productivity benchmarks show the 9950X3D2 post significant single-digit percentage gains over the 9950X3D. AMD did not announce price, but said that the 9950X3D2 Dual Edition will be available from April 22, 2026.
Known specifications of the Ryzen 9 9950X3D2 Dual Edition follow.

Add your own comment

149 Comments on AMD Announces Ryzen 9 9950X3D2 Dual Edition Desktop Processor with 192 MB L3 Cache

#126
MondoMike
Well that's a gamer and general audience pass up CPU, should be interesting if it can outpace the 9850x3d in gaming but with a price tag of $400+ more. X3D's are already wildly priced, 7800x3d was the only real option with a decent price tag before they bumped it up $100 a year or so ago.
Posted on Reply
#128
NoiseboxIT
As someone fortunate enough to be using multiple Ryzen 9 9950X3D's and who is building a living room SteamBox of sorts with a vastly cheaper Ryzen 9 9900X ($314 on Amazon from the Spring Sale)....I am tempted to get this chip, but the question is why? Like...I will be getting Nova Lake's biggest bLLC 52c baddie yes, but I hate the idea of a flagship chip right before a whole new generation is coming out. I was thrilled when Intel nuked the Core Ultra 9 Plus 290K because it also made no sense when Nova Lake was right around the corner. This is for sure the chip to beat...but will a standard Ryzen 9 10950 be 10% faster still for half the price? It's a toss up here....buy this and ride out most of Zen6 because how much better can it possibly be then this chip or just save and wait for Zen6 and then go harder?
Posted on Reply
#129
Sol_Badguy
NoiseboxITThis is for sure the chip to beat...but will a standard Ryzen 9 10950 be 10% faster still for half the price? It's a toss up here....buy this and ride out most of Zen6 because how much better can it possibly be then this chip or just save and wait for Zen6 and then go harder?
If the 10950X has 2x12 cores + HT even lacking X3D it will beat any prior Ryzen in MT. In ST also due to sky high clocks, especially in apps which scale less with X3D. In gaming... maybe gets a tie with Zen5 X3D.
Half the price of the 9950X3D2? Lol. It will likely keep the 9950X MSRP which was $650 before the $50 price cut a few days ahead of the Arrow Lake launch.

There is no dilemma here, everyone should wait for the new gens, as they can buy older gens then too. But that's after they see the uplift and how the Intel vs AMD variable turns out, then the older gens will get adjusted in price, just like Zen4 was relative to Zen5.
Posted on Reply
#130
A Computer Guy
Sol_BadguyIf the 10950X has 2x12 cores + HT even lacking X3D it will beat any prior Ryzen in MT. In ST also due to sky high clocks, especially in apps which scale less with X3D. In gaming... maybe gets a tie with Zen5 X3D.
Half the price of the 9950X3D2? Lol. It will likely keep the 9950X MSRP which was $650 before the $50 price cut a few days ahead of the Arrow Lake launch.

There is no dilemma here, everyone should wait for the new gens, as they can buy older gens then too. But that's after they see the uplift and how the Intel vs AMD variable turns out, then the older gens will get adjusted in price, just like Zen4 was relative to Zen5.
I'm hoping with the release of Intel Core Ultra 7 270K Plus AMD adjusts prices sooner. If Intel does release that for $300 that's a huge deal if someone wants 9950X level of performance and that might help hopefully drive down 9950X / 9950X3D / 9950X3D2 prices as well.
Posted on Reply
#131
AusWolf
NoiseboxITAs someone fortunate enough to be using multiple Ryzen 9 9950X3D's and who is building a living room SteamBox of sorts with a vastly cheaper Ryzen 9 9900X ($314 on Amazon from the Spring Sale)....I am tempted to get this chip, but the question is why? Like...I will be getting Nova Lake's biggest bLLC 52c baddie yes, but I hate the idea of a flagship chip right before a whole new generation is coming out. I was thrilled when Intel nuked the Core Ultra 9 Plus 290K because it also made no sense when Nova Lake was right around the corner. This is for sure the chip to beat...but will a standard Ryzen 9 10950 be 10% faster still for half the price? It's a toss up here....buy this and ride out most of Zen6 because how much better can it possibly be then this chip or just save and wait for Zen6 and then go harder?
I've already decided to wait for Zen 6. The 12 core CCD will be a much bigger improvement than X3D cache on the second CCD that games don't even use, and even if they do, they do so with extreme latency through the IF. Not to mention if Zen 6 also upgrades the IO die with higher memory speed support.
Posted on Reply
#132
efikkan
NoiseboxITAs someone fortunate enough to be using multiple Ryzen 9 9950X3D's and who is building a living room SteamBox of sorts with a vastly cheaper Ryzen 9 9900X ($314 on Amazon from the Spring Sale)....I am tempted to get this chip, but the question is why?
Yeah, why? How many do you need? :P
NoiseboxIT...but will a standard Ryzen 9 10950 be 10% faster still for half the price? It's a toss up here....buy this and ride out most of Zen6 because how much better can it possibly be then this chip or just save and wait for Zen6 and then go harder?
So you're concerned about the value aspect while considering upgrading something you probably don't need? ;)
We can only speculate about performance gains, and especially since it never translates directly to CPU performance anyways. For gaming, you'll probably see some that scale better with more computational power and some which are more bloated and cache sensitive. (Applications will scale much more linearly.)

None of us knows the performance characteristics of either Zen 6 or Nova Lake yet. But one thing of note is that Zen 5 added a lot on the execution side without a substantial performance gain as a result, which leads to the deduction that optimizing this might lead to a larger performance gain this time around. So I'm hoping for more than 10% this time, but then again AMD has said more about Zen 7 than Zen 6 if that's an indication...
AusWolfI've already decided to wait for Zen 6. The 12 core CCD will be a much bigger improvement than X3D cache on the second CCD that games don't even use, and even if they do, they do so with extreme latency through the IF. Not to mention if Zen 6 also upgrades the IO die with higher memory speed support.
It's not about using it or not, it's about where the point of diminishing returns is. Any cache line of instruction or data will be pushed to L3 when discarded from L2. Having two cache dies just means you have twice the amount of cache banks to spread the data across.

If we first assume you could add extra cache at the exact same latency, you'll get diminishing returns and some point where the gains just get negligible, depending on each workload.

Now, let's assume you have a 9950X3D and 9950X3D2 which managed to achieve the exact same clocks under any kind of load (they don't, but bear with me), the only real disadvantage I can think of with the second cache die having higher latency from the other CCD. I assume that all cache lines will be distributed by address like cache banks normally are, so it means that 50% of your L3 cache hits will be slightly slower while you get a slightly higher chance of a hit in L3 (not twice as large!), so you may get an intersection point where the benefits of the second cache die is outweighed by the disadvantage. (And probably a few where it doesn't matter either way.) Now consider that 9950X3D2 will actually be running at slightly lower clocks, which means many workloads will likely lose a little performance due to that too.
Posted on Reply
#133
Sol_Badguy
Following my estimation done in post #98 I will try to assess the following gains:
www.techpowerup.com/img/cOCQFUiRPWKnangg.jpg
Obviously the workload itself is essential as not every action in a certain program will produce the same difference between various CPUs.

In Blender the difference between the 9950X and 9950X3D is 6.07%. I suspect it's not the same test as in AMD's marketing slide.
tpucdn.com/review/intel-core-ultra-7-270k-plus/images/blender.png
But that 7% going from the 9950X3D to 9950X3D2 is almost like perfect scaling and with extra TDP on top.
www.tomshardware.com/pc-components/cpus/amd-ryzen-9-9950x3d-review/3
^Here it's basically 0% difference in the three Blender benchmarks.

V-Ray shows 7.79% from 9950X to 9950X3D.
tpucdn.com/review/intel-core-ultra-7-270k-plus/images/vray.png
Slide shows 7% from 9950X3D to 9950X3D2.
cdn.mos.cms.futurecdn.net/hZBDMhhzTmc6vPQJ5WRziH-970-80.png.webp
^Here it's just 0.61% difference.

Chromium compile shows 5% from 9950X3D to 9950X3D2.
tpucdn.com/review/intel-core-ultra-7-270k-plus/images/compiler.png
tpucdn.com/review/intel-core-ultra-7-270k-plus/images/compiler-react.png
MVS C++ and Java React are way less, 1.11% and 0.48% when going from 9950X to 9950X3D.
But okay not the same workload.

Unreal Engine compile shows 8% from 9950X3D to 9950X3D2.
tpucdn.com/review/intel-core-ultra-7-270k-plus/images/unreal-engine-5.png
^Here it's 6.94% from 9950X to 9950X3D.

Puget Bench for DaVinci Resolve is 7% from 9950X3D to 9950X3D2 in the slide.
Older data, that's what I found with both CPUs:
www.pugetsystems.com/labs/articles/amd-ryzen-9-9950x3d-and-9900x3d-content-creation-review/#Video_Editing_Motion_Graphics_DaVinci_Resolve_Studio
9950X3D is 3.8% better in overall score than 9950X.
cdn.mos.cms.futurecdn.net/b3aiHpFsTUKasMtzip2p2A-970-80.png.webp
^Here it's 1.99% difference.

In Geekbench MT 5.08% from 9950X to 9950X3D.
cdn.mos.cms.futurecdn.net/6pTpBsCxaoAiZCFBEgr2JA-970-80.png.webp
Matches the slide perfectly which implies almost perfect scaling when accounting for the extra TDP.

ONNX INT8 shows 6.29% from 9950X to 9950X3D.
cdn.mos.cms.futurecdn.net/9xMgPn3ARjGF8QZgXDiuHA-970-80.png.webp
In the slide we have 7% for ONNX inference workload.

I suppose AMD cherry-picked some tests where the gains are slide worthy so to speak.
But they still seem very optimistic with pretty much equal gains as those from 9950X to 9950X3D.
We'll see, but bang per buck is shaping up to be horrible now that the MSRP is known.
Posted on Reply
#134
AusWolf
Sol_BadguyFollowing my estimation done in post #98 I will try to assess the following gains:
www.techpowerup.com/img/cOCQFUiRPWKnangg.jpg
Obviously the workload itself is essential as not every action in a certain program will produce the same difference between various CPUs.

In Blender the difference between the 9950X and 9950X3D is 6.07%. I suspect it's not the same test as in AMD's marketing slide.
tpucdn.com/review/intel-core-ultra-7-270k-plus/images/blender.png
But that 7% going from the 9950X3D to 9950X3D2 is almost like perfect scaling and with extra TDP on top.
www.tomshardware.com/pc-components/cpus/amd-ryzen-9-9950x3d-review/3
^Here it's basically 0% difference in the three Blender benchmarks.

V-Ray shows 7.79% from 9950X to 9950X3D.
tpucdn.com/review/intel-core-ultra-7-270k-plus/images/vray.png
Slide shows 7% from 9950X3D to 9950X3D2.
cdn.mos.cms.futurecdn.net/hZBDMhhzTmc6vPQJ5WRziH-970-80.png.webp
^Here it's just 0.61% difference.

Chromium compile shows 5% from 9950X3D to 9950X3D2.
tpucdn.com/review/intel-core-ultra-7-270k-plus/images/compiler.png
tpucdn.com/review/intel-core-ultra-7-270k-plus/images/compiler-react.png
MVS C++ and Java React are way less, 1.11% and 0.48% when going from 9950X to 9950X3D.
But okay not the same workload.

Unreal Engine compile shows 8% from 9950X3D to 9950X3D2.
tpucdn.com/review/intel-core-ultra-7-270k-plus/images/unreal-engine-5.png
^Here it's 6.94% from 9950X to 9950X3D.

Puget Bench for DaVinci Resolve is 7% from 9950X3D to 9950X3D2 in the slide.
Older data, that's what I found with both CPUs:
www.pugetsystems.com/labs/articles/amd-ryzen-9-9950x3d-and-9900x3d-content-creation-review/#Video_Editing_Motion_Graphics_DaVinci_Resolve_Studio
9950X3D is 3.8% better in overall score than 9950X.
cdn.mos.cms.futurecdn.net/b3aiHpFsTUKasMtzip2p2A-970-80.png.webp
^Here it's 1.99% difference.

In Geekbench MT 5.08% from 9950X to 9950X3D.
cdn.mos.cms.futurecdn.net/6pTpBsCxaoAiZCFBEgr2JA-970-80.png.webp
Matches the slide perfectly which implies almost perfect scaling when accounting for the extra TDP.

ONNX INT8 shows 6.29% from 9950X to 9950X3D.
cdn.mos.cms.futurecdn.net/9xMgPn3ARjGF8QZgXDiuHA-970-80.png.webp
In the slide we have 7% for ONNX inference workload.

I suppose AMD cherry-picked some tests where the gains are slide worthy so to speak.
But they still seem very optimistic with pretty much equal gains as those from 9950X to 9950X3D.
We'll see, but bang per buck is shaping up to be horrible now that the MSRP is known.
Bang per buck was probably never a consideration anyway. The 7800X3D and 7700 non-X are still kings in that regard (when it comes to gaming at least).
Posted on Reply
#135
Sol_Badguy
@AusWolf

True but it still feels like AMD is exaggerating here with the MSRP.
9950X3D was $700, up $100 from 9950X's revised MSRP of $600 (down from $650 two months prior).
So this should've been $800 tops.
It's like they're putting an extra Zen5 flagship tax on top.
That flagship tax could've been (partially) justified somewhere in July-August 2025, but now it's so late (with Zen6 probably less than a year away) that it actually feels like an IQ test instead.
But whatever, not my problem.
Posted on Reply
#136
NoiseboxIT
Sol_Badguy@AusWolf

True but it still feels like AMD is exaggerating here with the MSRP.
9950X3D was $700, up $100 from 9950X's revised MSRP of $600 (down from $650 two months prior).
So this should've been $800 tops.
It's like they're putting an extra Zen5 flagship tax on top.
That flagship tax could've been (partially) justified somewhere in July-August 2025, but now it's so late (with Zen7 probably less than a year away) that it actually feels like an IQ test instead.
But whatever, not my problem.
I don’t think Zen7 is anywhere near. Zen6 will launch at CES upcoming basically (especially Zen6X3D). Zen7 will likely be 18 months after that. Zen5 has had quite a long run as is, 18 months come Zen6 releasing.

Remember, Zen7 won’t be with us until at least Q3-Q4 2028.
Posted on Reply
#137
NoiseboxIT
AusWolfBang per buck was probably never a consideration anyway. The 7800X3D and 7700 non-X are still kings in that regard (when it comes to gaming at least).
I feel like I will skip this chip. It just doesn’t bring much at all to the party. $800 for a preview of Nova Lake is not worth it. Especially when we all know a basic Zen6 Ryzen 9 10950X will outperform it even sans 3D V-Cache.

It’s admirable AMD even launched this, but essentially it’s an engineering proof of concept and it meant for the real world. Nova Lake is the elephant in the room, especially now that the lower end chip is 44 cores overall and the high end one 52c.

It’s good to see competition again, but while Zen5 is indeed a good place to be, even Intel got it right cancelling the Core Ultra 9 Plus 290K. It would have been the same weirdness. A new flagship right before a whole new generation? Like seriously, who’s buying that?? No one. Hence cancelled. AMD should have skipped this chip entirely. It’s very likely the 12 core single CCD Ryzen 9 10900X will have similar performance for half the price.
Posted on Reply
#138
AusWolf
Sol_Badguy@AusWolf

True but it still feels like AMD is exaggerating here with the MSRP.
9950X3D was $700, up $100 from 9950X's revised MSRP of $600 (down from $650 two months prior).
So this should've been $800 tops.
It's like they're putting an extra Zen5 flagship tax on top.
That flagship tax could've been (partially) justified somewhere in July-August 2025, but now it's so late (with Zen7 probably less than a year away) that it actually feels like an IQ test instead.
But whatever, not my problem.
Don't get me wrong, I agree completely. Flagship products are rarely meant for value conscious buyers.

And indeed, it does feel like a last push before the Zen 6 launch to milk the current architecture as much as they can, not unlike Nvidia's Super cards.
Posted on Reply
#139
Sol_Badguy
AusWolfAnd indeed, it does feel like a last push before the Zen 6 launch to milk the current architecture as much as they can, not unlike Nvidia's Super cards.
Well the 40 series Supers (4070 Super and 4070 Ti Super) did have a $50 discount so with all that greed nVidia still managed some charity.
This could've changed with 50 Supers though.

Intel also did some hardcore charity with Arrow Lake Refresh (Jacketman is paying so why not?).

But AMD wants to play the "high roller strong milker not one step back" card.
Understandable as they probably won't get another such opportunity in the near future.
NoiseboxITI don’t think Zen7 is anywhere near. Zen6 will launch at CES upcoming basically (especially Zen6X3D). Zen7 will likely be 18 months after that. Zen5 has had quite a long run as is, 18 months come Zen6 releasing.

Remember, Zen7 won’t be with us until at least Q3-Q4 2028.
Lol @ goddamn typo, I've corrected the post, Zen 6 that's what I meant.
Posted on Reply
#140
AusWolf
Sol_BadguyWell the 40 series Supers (4070 Super and 4070 Ti Super) did have a $50 discount so with all that greed nVidia still managed some charity.
This could've changed with 50 Supers though.
That's hardly charity when the base product was way overpriced to begin with, but anyway, you get my point.
Posted on Reply
#141
Sol_Badguy
@AusWolf

Of course.
The best spin was with the 4080 which had a candid camera $1200 MSRP which then could be reduced to all sorts of degrees and appear like charity while still creating a new overpriced baseline for the future. :roll:

Granted one year later the 5070 Ti managed to offer 4080 Super level of performance ($1000) for 3/4 of the price.
So yeah not awesome but not horrible either.
Two years later the 6070 will do the same for probably $550 (fingers crossed). And that's without adjusting for inflation.
Posted on Reply
#142
AusWolf
Sol_BadguyTwo years later the 6070 will do the same for probably $550 (fingers crossed). And that's without adjusting for inflation.
I see no chance of that happening, but let's not derail the topic any further. :)
Posted on Reply
#143
jesdals
The first batch of test seems to indicate that unfortunately the value is very poor on the cpu. Unless You have specific workloads that need the cache - an overall choise would rather be 9800/9850x3d for gameing and 9950x3d for mixed use
Posted on Reply
#144
AVATARAT
From the first two reviews (der8auer and HU), it seems that so far no one has compared the 9950x3D2 to the 9850x3D :roll:
But from what I've seen, it performs better than I expected, likely due to its higher clock speed.

Posted on Reply
#145
Pets4Ever
Has anyone tested the edge cases in gaming yet? Like for example, BeamNG.drive
Posted on Reply
#146
AusWolf
jesdalsThe first batch of test seems to indicate that unfortunately the value is very poor on the cpu. Unless You have specific workloads that need the cache - an overall choise would rather be 9800/9850x3d for gameing and 9950x3d for mixed use
As expected. Anyone who ever tried a dual CCD CPU knows that it's not the lack of cache that holds them back, but the slow inter-CCD communication through the infinity fabric. They're very good at parallel workloads, but as soon as you give them a single job that works across more than one CCD (like game shader caching), they die. You're much better off with a single CCD X3D CPU at that point.
Posted on Reply
#147
igormp
AusWolfbut as soon as you give them a single job that works across more than one CCD (like game shader caching), they die
Minor nit, but I dont think cached shaders are a good example since you just submit those to the GPU and that's it.
The issue with games is that the threads in a game loop require lots of communication with one another, so that's where the inter-CCD latency kills the performance for those and won't allow them to scale properly. Think of a physics thread trying to relay data to the thread that cares care of audio.
Other kinds or parallel workloads that can go by with no or close to no communication across threads tend to scale really well without facing such issues.

But nonetheless, this CPU shows that the extra CCD doesn't really help with games in a superficial manner and that the extra clocks are what brought most of the benefits for that scenario. From Phoronix's review, it had some really nice uplifts to HPC and other memory intensive tasks, but those were expected already.

Ofc if you have the money this is faster than the other models, has no downsides (other than the extra power consumption), and makes it so one doesn't need to care about the whole CCD/no-CCD scheduling.
Posted on Reply
#149
A Computer Guy
AVATARATIn some of Phoronix's benchmarks, the cache helps significantly, so I guess AMD has, as usual, shot itself in the leg, i.e., its TR market...
www.phoronix.com/review/amd-ryzen-9950x3d2-linux
Well not so much. Remember TR is a platform with significantly more bus and I/O capabilities over AM5. Multiple GPU's, big storage, big ram, more cores, little to no lane sharing, better feature validation, IPMI, etc... While it gets stripped out in performance at 16 core level by newer AM5 chips, AM5 is severely limited while also much cheaper.
Posted on Reply
Add your own comment
Nov 22nd, 2026 11:01 CST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts