News Russia develops its first lithography tool — outdated by 30 years from day one

[Admin]

Russia can now build lithography system capable of making 350nm chips.

Russia develops its first lithography tool — outdated by 30 years from day one : Read more

 

[peachpuff]

"We assembled and made the first domestic lithographic scanner,"

Found a picture of it...

 

[Pierce2623]

Found a picture of it...

Nope

 

[JamesSneed]

350nm = 0.350 micron aka 0.35 µm

 

[ivan_vy]

not bad for a country with 1/13th GPD of USA, still too old node to be useful. 350nm is the node Pentium -Pentium II was made.
I think is more like a proof of concept and to training engineers, for everything else they'll have to buy the chinese offerings.

 

[Eximo]

As the article states, you can make a lot of useful, and absolutely necessary, electronics with larger nodes. And keep in mind that a lot of military equipment in use today was made well before the Pentium I.

 

[peachpuff]

Lol at the updated photo, Anton is trolling hard!

 

[The Historical Fidelity]

not bad for a country with 1/13th GPD of USA, still too old node to be useful. 350nm is the node Pentium -Pentium II was made.
I think is more like a proof of concept and to training engineers, for everything else they'll have to buy the chinese offerings.

Well better late than never I guess. I believe they can manufacture 250nm using multi patterning or even a half node beyond (220nm) with quad patterning. Heck, if SMIC shares their etching technique that allows them to make 5nm class features, perhaps Russia can eke out a 180nm process. It’ll be interesting to see what they can do bringing modern techniques to old lithography technology.

 

[blargh4]

The world of semiconductors is broader than smartphone SoCs and data center GPUs; there is a heap of silicon in current production that uses 180nm+ lithography. I suspect Russia's priorities lie in beefing up homegrown capabilities for national security related concerns first and foremost, competing with ASML probably isn't too high on that list. They already have domestic production capability for 65nm chips, though clearly not using domestic tools.

 

[LabRat 891]

As I recall, larger lithography chips are much-less liable to be damaged by ionizing radiation, etc.

Kinda reminds me of the post-war technological recovery in the CyberPunk setting.
Anything mid-war and immediately post-war, was 'primitive' on purpose.

 

[dlheliski]

Igor, please pass crayon, I need to do M3 layer.

 

[tbq]

350nm is small enough for most applications and is used in far more products than most tech savvy people probably realize. Shrinking a process node (or a dozen process nodes) has no impact on the performance on most of the chips made with these nodes. Your microwave won't heat your food faster using a 7nm controller compared to a 600nm controller. Virtually anything in your house that plugs into the wall or uses batteries has some sort of chip inside and rarely does the lithography generation make any difference of its functionality.

 

[Amdlova]

I can't overclock my oven? Lol 🙁

 

[cyrusfox]

Commendable effort, on their way to establish independent capability, bleeding edge is not needed for majority of use cases for microelectronics.

Its too bad that ASML is the only horse in the race at the bleeding edge, The improvements from here for ASML are very slow, So there is time for newcomers to come in and iterate to be a cost competitive solution as well as pick up the rest of the market. Litho is expensive, competition is welcome.

 

[umeng2002_2]

I applaud every country to become as economically independent as possible.

 

[brandonjclark]

I wonder how many of you are impressed by this "publicly-funded" effort to create it's first tooling as much as I am?

There were PRACTICAL lessons learned around the processes of building tooling like this that needed to be DISCOVERED and DEVELOPED for the first time.

Could they have shot for a smaller process node? I suppose, maybe.

But I think, whether they had a choice or not, it is beneficial to start with something "easier" just so you can focus on the steps needed to get from a thought to a real product.

All that has been figured out now.

Now it's just a matter of refining those steps, and they will.

Hats off to a country for learning!

 

[shady28]

The world of semiconductors is broader than smartphone SoCs and data center GPUs; there is a heap of silicon in current production that uses 180nm+ lithography. I suspect Russia's priorities lie in beefing up homegrown capabilities for national security related concerns first and foremost, competing with ASML probably isn't too high on that list. They already have domestic production capability for 65nm chips, though clearly not using domestic tools.

Exactly. UMC, Taiwan's 2nd largest semi maker and the one that makes critical automotive chips, only has one fab out of 11 that can do 28 / 14 nm. Most of the rest are 150nm or larger. Those are what run cars, trucks, smart AC systems, etc.

I'd bet they'll have 250nm within a year or so. That's P3 and the old DEC Alpha node.

In reality, what this does is give them the ability to make those supporting chips for relatively mundane tasks domestically, while heavier hitting SoCs and CPUs can be made in China on SMICs 14-28nm nodes.

 

[bobba84]

Ahahahahah! I said a few weeks ago they are going to restart making the Soviet 8088 clones, I wasn't far off!

 

[serget]

How many countries can produce their own chips? 5? 10? How many of them do have lithography?.. so, not too bad to have even 30 years old technologies here. They will improve it pretty quickly.

 

[Midav]

not bad for a country with 1/13th GPD of USA, still too old node to be useful. 350nm is the node Pentium -Pentium II was made.
I think is more like a proof of concept and to training engineers, for everything else they'll have to buy the chinese offerings.

Russia has significant expertese in plasma physics, the crux of EUV. I exp

 

[COLGeek]

Please stick to the tech topic and not politics. Thank you.

 

[The Historical Fidelity]

Russia has significant expertese in plasma physics, the crux of EUV. I exp

There are many more variables that are arguably harder to master than plasma when it comes to EUV lithography.