Thursday, December 11th 2025

Chinese SMIC Achieves 5 nm Production on N+3 Node Without EUV Tools

Chinese company SMIC has officially achieved volume production of its newest 5 nm-class node called SMIC N+3. This is officially China's most advanced semiconductor node produced without any extreme ultraviolet (EUV) lithography tools, relying on the deep ultraviolet (DUV) to manufacture its silicon. In the latest analysis of the Huawei Kirin 9030 SoC, the TechInsights has confirmed that the chip has officially been manufactured using the new advanced node, bringing Chinese semiconductor independence one step closer to reality. This node is a full generation ahead of the older SMIC N+2, which is 7 nm-class node Huawei has been using to manufacture its Ascend series of AI accelerators and other infrastructure parts.

However, this is not without drawbacks. Since SMIC uses DUV to print silicon designs, the company faces significant challenges in realizing its silicon ambitions. For printing smaller nodes, EUV scanners offer more possibilities due to their much smaller wavelength of 13.5 nm, compared to the smallest DUV immersion scanner wavelength of 193 nm. TechInsights has confirmed that SMIC's N+3 node, despite achieving impressive DUV multi-patterning implementation, encounters significant yield challenges, particularly due to the aggressively scaled metal pitch. As a result, the Huawei Kirin 9030 SoC is likely produced at an operating loss, with a significant portion of dies being discarded or used for downgraded chips.
In a single-patterning round, the best DUV immersion scanners can achieve resolutions for nodes with sub-38 nm prints. SMIC has likely optimized these machines to reach closer to 35 nm in a single print, using multiple rounds of printing to fully etch the design. Techniques like self-aligned quadruple patterning (SAQP) have been used in semiconductor nodes for years, so it is logical that SMIC has employed a similar technique with its new N+3 node. It is important to note that the majority of these DUV tools has been studied for decades, and new techniques are just refinements of what has been internally applied at test runs. TechInsights expects that the yield issues are still present, but information is limited about SMIC's node success.

In September, SMIC was testing an immersion DUV lithography scanner developed by Shanghai Yuliangsheng Technology. This move is part of an effort to establish a local wafer fabrication equipment industry and lessen reliance on Western suppliers. The Yuliangsheng system was described as an immersion DUV scanner intended for 28 nm class processes, comparable to older Twinscan designs from around 2008. However, it isn't possible that SMIC has achieved a breakthrough in such a short period, and the new DUV scanner might be domestically producing Huawei's Kirin 9030 SoC. Hence, this new SMIC N+3 node is likely made using ASML's tools.
Source: TechInsights
Add your own comment

18 Comments on Chinese SMIC Achieves 5 nm Production on N+3 Node Without EUV Tools

#1
vbq7qK68eyYAH4iR
Mere Mortal
This was the same when they announced they could do 7nm with DUV. Yields were bad but they could do it. If they can't get access to EUV, they're doing the best they can and are showing the world they can do it.
Posted on Reply
#2
Panther_Seraphin
You CAN go really small on DUV, its just going to economically suicidal due to yields/time/energy costs.
Posted on Reply
#3
freeagent
That is pretty impressive no matter how you look at it..

If I gave you a hammer and sent you off into the woods.. how long till you can send me an email?
Posted on Reply
#4
zzd911
There's an idiom "achieve half the result with twice the effort".
事倍功半。
Posted on Reply
#5
SemperPC09
This is impressive. Like they say, "where there is a will, there is a way." Very cool stuff. It is just another reminder that you don't need the best or the newest thing to do something big. I wonder what yields are though and cost.
Posted on Reply
#6
Chaitanya
SemperPC09This is impressive. Like they say, "where there is a will, there is a way." Very cool stuff. It is just another reminder that you don't need the best or the newest thing to do something big. I wonder what yields are though and cost.
You need to see the inovation Chinese Torch makers are puting in their lights and LEDs(Cree, Osram, Phillips, Samsung, Luminous and Nichia are falling way behind the curve at this point).
flashlight.nitecore.com/Know/detail/ad/1875
flashlight.nitecore.com/Know/detail/ad/2022
flashlight.nitecore.com/Know/detail/ad/2108
www.firefly-outdoor.com/collections/emitter-and-battery
Skilhunt also has their own homegrown LED(HighCRI in 5k Tint) but they dont have dedicated page about that LED.
Posted on Reply
#7
SemperPC09
ChaitanyaYou need to see the inovation Chinese Torch makers are puting in their lights and LEDs(Cree, Osram, Phillips, Samsung, Luminous and Nichia are falling way behind the curve at this point).
flashlight.nitecore.com/Know/detail/ad/1875
flashlight.nitecore.com/Know/detail/ad/2022
flashlight.nitecore.com/Know/detail/ad/2108
www.firefly-outdoor.com/collections/emitter-and-battery
Skilhunt also has their own homegrown LED(HighCRI in 5k Tint) but they dont have dedicated page about that LED.
Thanks for sharing! I did not know how far they had come.

Really impressive stuff they are doing even without access to the latest and greatest. It also goes to show how far behind and how slow other companies (that have advanced equipment) are. Ingenuity drives advancements. Kudus to the brilliant minds in those companies without the luxury of the best equipment.
Posted on Reply
#8
john_
People overclocking CPUs, GPUs and RAM, SMIC overclocking manufacturing nodes.
Posted on Reply
#9
TumbleGeorge
Hmm in 2027 mass production with chinese own EUV instruments? They already have developed the first one.
Posted on Reply
#10
Vayra86
This 'race' feels rather similar to the AI race where the tooling is clearly not up to the task and the result is a highly cost intensive operation to produce something we already have.

If someone can explain to me what the point is of having a way overpriced 5nm chip relative to a way overpriced 7nm chip, I'm all ears.

This is a prestige project and a loss venture. At best, it is mildly useful in military applications where cost is no object. On a marketplace though, its not just dead, its an exercise in self destruction. You simply cannot mass produce this and get away with it for too long.
TumbleGeorgeHmm in 2027 mass production with chinese own EUV instruments? They already have developed the first one.
Dream on :D
Posted on Reply
#11
londiste
Aggressively scaled metal pitch and SAQP sounds eerily reminiscent of Intel-s 7nm woes.
Posted on Reply
#12
rwat888
Panther_SeraphinYou CAN go really small on DUV, its just going to economically suicidal due to yields/time/energy costs.
US is spending trillions on data centers, if 5nm on DUV only costs +40-50% higher than 7nm on DUV, then that is chump change bro. For reference, 2nm costs +50% higher cost than 3nm, that's just the name of the game. For China on AI GPUs, 5nm is more than sufficient for years.
Posted on Reply
#13
W1zzard
Vayra86If someone can explain to me what the point is of having a way overpriced 5nm chip relative to a way overpriced 7nm chip, I'm all ears.
Security and Sovereignty, both supported with infinite money by all governments
Posted on Reply
#14
igormp
Vayra86If someone can explain to me what the point is of having a way overpriced 5nm chip relative to a way overpriced 7nm chip, I'm all ears.
Aside from what was said above, the end results are also often more efficient, meaning and you can get more performance at the same power draw, or keep the same performance at a lower power draw.
A more expensive chip is a one-time cost, but its power draw is recurrent and thus a more efficient chip can make up for its initial price in the medium term, specially when we're talking about larger scales and how electricity is a limited and expensive commodity.
Posted on Reply
#15
Smartcom5
Panther_SeraphinYou CAN go really small on DUV …
Yup, getting down to 7nm-density levels using DUVL exclusively, was known to be possible from like 2014 or 2016.
Now, as far as I'm aware, it's possible to reach process-metrics as small as down to 1.4 Ångström (1.4nm) or so through DUVL only (using multi-patterning, obviously).
Posted on Reply
#16
Wirko
W1zzardSecurity and Sovereignty, both supported with infinite money by all governments
And concentration of knowledge, which they can build upon at any later time, when conditions (technology, markets, politics, natural resources) become more favourable.
Posted on Reply
#17
Panther_Seraphin
Smartcom5Yup, getting down to 7nm-density levels using DUVL exclusively, was known to be possible from like 2014 or 2016.
Now, as far as I'm aware, it's possible to reach process-metrics as small as down to 1.4 Ångström (1.4nm) or so through DUVL only (using multi-patterning, obviously).
Yeah thats the thing, you CAN get down there but the amount of passes with multi patterning to get down there vs doing it with a single pattern EUV means from a scale perspective EUV wins in a consumer market. Especially when you consider each pass can introduce defects/anomalies so doing the 16 or so passes to get down there inherintley leads to abysmal yields.

As Wizard pointed out though from a soverignty/security argument that throws the commercial argument completely on its head.
Posted on Reply
#18
awdrifter
zzd911There's an idiom "achieve half the result with twice the effort".
事倍功半。
Well, they don't have access to EUV lithography machines, so they have to use DUV. This is more like 弯道超车 (overtake during the corner).
Posted on Reply
Add your own comment
Nov 19th, 2026 18:54 CST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts