You stole this. HBM3E. The patents for this belongs to netlist and you don’t have a license why your court starts next week. Can’t wait for the injunction. In Europe courts already are putting together an injunction for a cease a desist. 😂😂😂 good luck in selling stolen products
Good time to invest in Micron and nVidia. Only question is when the latter bursts because like Cisco 25 years ago, their products aren't entirely unique - they just have a head start over everyone else.
Hynix is years away from the volume Micron has. They're behind on Yongin Semiconductor, Fab 1 won't be online until next year, and who knows when Fab 2-4 will be complete. Nice try tho!
More specifically (and a google search will back this up) Micron has 26% of the DRAM market while Hynix has 30% of the DRAM market. But within that market, Micron has overtaken Hynix on HBM3 since last year, and while numbers are unavailable because we are talking about a new product that is barely shipping, the fact Micron has allotted its entire supply of HBM3e, and is shipping, while Hynix just started production 6 days ago and wont ship for possibly months, there are two key takeaways here:
1) Micron has more partner confidence as they were sampling over a year ago 2) Micron is actively shipping product
Micron does not have 1/3 of the HBM production volume that SK Hynix has. You just looked at the specs and thought that translates into market dominance. Both volume and yield are lower.
SK Hynix has always been the dominant player in the HBM market, no wonder it was chosen to help get the technology off the ground. At the moment it has 80% of the market.
Forgetting the fact a Google search proves you are full of shit, just think logically here: nVidia, the third most valuable private company on the planet, has exclusively partnered with a company which, according to you, has 20% of the market for a component nVidia needs for their SURVIVAL in AI?
I might be going out on a limb here but I'd bet you don't know a lot about operating a successful business.
Nvidia and AMD will simply use whatever is available, that doesn't mean anything. I doubt Micron has 20% of the market, if they did it would be a huge victory for them.
Micron doesn't have the potential production volume for HBM3 of Hynix or Samsung, but as the later two companies have yet to ship anything, at the moment, Micron has the entire pie, and assuming Hynix and Samsung ship on time, the average for the fiscal year will put Micron ahead, even if Hynix ships a higher volume they would have to do it at an exponential rate by Fall.
We are talking about a complex product. SK Hynix has a special technology that allows it to reap the highest yield among the three HBM manufacturers, which is why it has 80% of the market in its hands.
"One of the reasons Samsung has fallen behind is its decision to stick with chip making technology called non-conductive film (NCF) that causes some production issues, while Hynix switched to the mass reflow molded underfill (MR-MUF) method to address NCF's weakness, according to analysts and industry watchers."
People love just making shit up around here, like they know the inner workings of these companies yields - literally trade secrets - and production capacity for a product that's barely been in volume production for a few months. Hynix just started volume production on March 19th 2024
SKHynix hasn't had 50% HBM marketshare in any fiscal year, ever. They have split majority marketshare with Samsung, with Micron in third - until now. Micron at the moment has 100% marketshare because they are the only one that has shipped anything. While that will change, at the moment, SKHynix and Samsung have 0% of the HBM3e market because they are still in production.
It's not the A100/H100 that this HBM is being sold for, it's for a high end GPU for us... Yes, that's what it is... And this soon to be high end GPU is going to be sold at a reasonable price... Yes, that's what's going to happen... I can see it all so clearly now...
Samsung produced 300 M to 500 M Ampere on channel supply data. Seems to have been a cost : margin agreement with a capped Nvidia supply commitment, Samsung was selling direct to AIBs? I think so.
Also suggests there were Ampere risk/ramp costs to overcome reported by many although I never fully bought into that on the volume of the full production run to achieve Samsungs margin objective.
From Ada mobile dGPU design generation to Blackwell desktop design generation seems reasonable moving back to wide bus + VRAM advantage on a depreciated 4 nm process trades off dGPU cost with board, trace and VRAM cost.
I believe BW RTX consumer will be more reasonably priced then many speculate. The reason is to appease game land and more important to encourage Ampere to Blackwell upgrade for which there is a hidden Nvidia agenda.
The agenda is a world market 'bargain' priced RT/Nvidia/basic ML/AI development platform; entry level on the CPUI instruction sets limits and Ampere features, serviced by SI integration relying on 300 M used Ampere attached to 300 M plus used Xeon v3/v4/Skylake. That development slingshot on my supply data in relation Nvidia applications segment stronghold position is actual the big BW priced right win, driving world market development onto Ampere attached to capable enough Xeon at a world market price. mb
I am wondering if given some more years (somewhere between 2026 and 2032), this steap increase in HBM volume manufacturing could help decrease enough of the HBM price premium versus GDDR memory, or even LPDDR memory to make it possible to be a premium option for the consumer market ?
It would then be possible to see some premium computing products (ex: Apple Mac Pro, Apple Studio, Apple Macbook Pro,…) have the option for 32GB / 64GB / 128GB HBM memory, and maybe a few years later on Macbook Air / iPad / iPhone as HBM memory seems more power efficient per bit than some other memory types…
Even better would be to have Non-Volatile-Memory (NVM) like SOT-MRAM memory dies be used to create Non-Volatile HBM stack : it would enable so many new opportunities !!! (Ex: a 32GB stack of low latency, high bandwith Non-Volatile-Memory / storage).
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
27 Comments
Back to Article
PeachNCream - Friday, March 22, 2024 - link
Typo error "...high-bandwidth memory or 2024..."Perhaps "or" should be "for"
Sonicsgp1 - Friday, March 22, 2024 - link
You stole this. HBM3E. The patents for this belongs to netlist and you don’t have a license why your court starts next week. Can’t wait for the injunction. In Europe courts already are putting together an injunction for a cease a desist. 😂😂😂 good luck in selling stolen productstrivik12 - Friday, March 22, 2024 - link
I think netlist is a patent troll and will win nothing.Samus - Friday, March 22, 2024 - link
Good time to invest in Micron and nVidia. Only question is when the latter bursts because like Cisco 25 years ago, their products aren't entirely unique - they just have a head start over everyone else.Dante Verizon - Friday, March 22, 2024 - link
You mean SK Hynix*Samus - Monday, March 25, 2024 - link
Hynix is years away from the volume Micron has. They're behind on Yongin Semiconductor, Fab 1 won't be online until next year, and who knows when Fab 2-4 will be complete. Nice try tho!Samus - Monday, March 25, 2024 - link
More specifically (and a google search will back this up) Micron has 26% of the DRAM market while Hynix has 30% of the DRAM market. But within that market, Micron has overtaken Hynix on HBM3 since last year, and while numbers are unavailable because we are talking about a new product that is barely shipping, the fact Micron has allotted its entire supply of HBM3e, and is shipping, while Hynix just started production 6 days ago and wont ship for possibly months, there are two key takeaways here:1) Micron has more partner confidence as they were sampling over a year ago
2) Micron is actively shipping product
Dante Verizon - Monday, March 25, 2024 - link
Micron does not have 1/3 of the HBM production volume that SK Hynix has. You just looked at the specs and thought that translates into market dominance. Both volume and yield are lower.SK Hynix has always been the dominant player in the HBM market, no wonder it was chosen to help get the technology off the ground. At the moment it has 80% of the market.
Samus - Wednesday, March 27, 2024 - link
Forgetting the fact a Google search proves you are full of shit, just think logically here: nVidia, the third most valuable private company on the planet, has exclusively partnered with a company which, according to you, has 20% of the market for a component nVidia needs for their SURVIVAL in AI?I might be going out on a limb here but I'd bet you don't know a lot about operating a successful business.
Dante Verizon - Wednesday, March 27, 2024 - link
Nvidia and AMD will simply use whatever is available, that doesn't mean anything. I doubt Micron has 20% of the market, if they did it would be a huge victory for them.BvOvO - Tuesday, April 9, 2024 - link
NVIDIA:"Company type Public"
Samus - Thursday, March 28, 2024 - link
Micron doesn't have the potential production volume for HBM3 of Hynix or Samsung, but as the later two companies have yet to ship anything, at the moment, Micron has the entire pie, and assuming Hynix and Samsung ship on time, the average for the fiscal year will put Micron ahead, even if Hynix ships a higher volume they would have to do it at an exponential rate by Fall.kkww - Wednesday, March 27, 2024 - link
Dante Verizon is correct and you are so wrong it's not even funny. Maybe you should learn how to google better.Samus - Thursday, March 28, 2024 - link
Maybe you should learn what business fiscal year production is.Dante Verizon - Friday, March 22, 2024 - link
Micron's HBM production volume and yields are quite limited.ballsystemlord - Friday, March 22, 2024 - link
Could you give a link for that claim? I'd like to read more on the matter. Why are their yields so bad?Terry_Craig - Saturday, March 23, 2024 - link
We are talking about a complex product. SK Hynix has a special technology that allows it to reap the highest yield among the three HBM manufacturers, which is why it has 80% of the market in its hands."One of the reasons Samsung has fallen behind is its decision to stick with chip making technology called non-conductive film (NCF) that causes some production issues, while Hynix switched to the mass reflow molded underfill (MR-MUF) method to address NCF's weakness, according to analysts and industry watchers."
Adramtech - Saturday, March 23, 2024 - link
That’s an unverified beliefSamus - Monday, March 25, 2024 - link
People love just making shit up around here, like they know the inner workings of these companies yields - literally trade secrets - and production capacity for a product that's barely been in volume production for a few months. Hynix just started volume production on March 19th 2024Dante Verizon - Monday, March 25, 2024 - link
Sk Hynix has 80% of the HBM market, foolThe Hardcard - Monday, March 25, 2024 - link
They won’t have 80 percent of 2024 production when year end results are released in early 2025.Samus - Thursday, March 28, 2024 - link
SKHynix hasn't had 50% HBM marketshare in any fiscal year, ever. They have split majority marketshare with Samsung, with Micron in third - until now. Micron at the moment has 100% marketshare because they are the only one that has shipped anything. While that will change, at the moment, SKHynix and Samsung have 0% of the HBM3e market because they are still in production.Terry_Craig - Saturday, March 23, 2024 - link
https://www.reuters.com/technology/samsung-use-chi...ballsystemlord - Friday, March 22, 2024 - link
It's not the A100/H100 that this HBM is being sold for, it's for a high end GPU for us... Yes, that's what it is... And this soon to be high end GPU is going to be sold at a reasonable price... Yes, that's what's going to happen... I can see it all so clearly now...Oxford Guy - Sunday, March 24, 2024 - link
And there will be good games instead of the same warmed-over cartoonish stuff.Bruzzone - Monday, March 25, 2024 - link
Samsung produced 300 M to 500 M Ampere on channel supply data. Seems to have been a cost : margin agreement with a capped Nvidia supply commitment, Samsung was selling direct to AIBs? I think so.Also suggests there were Ampere risk/ramp costs to overcome reported by many although I never fully bought into that on the volume of the full production run to achieve Samsungs margin objective.
From Ada mobile dGPU design generation to Blackwell desktop design generation seems reasonable moving back to wide bus + VRAM advantage on a depreciated 4 nm process trades off dGPU cost with board, trace and VRAM cost.
I believe BW RTX consumer will be more reasonably priced then many speculate. The reason is to appease game land and more important to encourage Ampere to Blackwell upgrade for which there is a hidden Nvidia agenda.
The agenda is a world market 'bargain' priced RT/Nvidia/basic ML/AI development platform; entry level on the CPUI instruction sets limits and Ampere features, serviced by SI integration relying on 300 M used Ampere attached to 300 M plus used Xeon v3/v4/Skylake. That development slingshot on my supply data in relation Nvidia applications segment stronghold position is actual the big BW priced right win, driving world market development onto Ampere attached to capable enough Xeon at a world market price. mb
Diogene7 - Monday, March 25, 2024 - link
I am wondering if given some more years (somewhere between 2026 and 2032), this steap increase in HBM volume manufacturing could help decrease enough of the HBM price premium versus GDDR memory, or even LPDDR memory to make it possible to be a premium option for the consumer market ?It would then be possible to see some premium computing products (ex: Apple Mac Pro, Apple Studio, Apple Macbook Pro,…) have the option for 32GB / 64GB / 128GB HBM memory, and maybe a few years later on Macbook Air / iPad / iPhone as HBM memory seems more power efficient per bit than some other memory types…
Even better would be to have Non-Volatile-Memory (NVM) like SOT-MRAM memory dies be used to create Non-Volatile HBM stack : it would enable so many new opportunities !!! (Ex: a 32GB stack of low latency, high bandwith Non-Volatile-Memory / storage).