[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]  Jukan [@Jukanlosreve](/creator/twitter/Jukanlosreve) on x 22.6K followers Created: 2025-07-23 23:39:04 UTC Report: Samsung Electronics' HBM4 Mass Production Expected 'Next Year'... Sample Test Yield at XX% Samsung Electronics' mass production of 12-stack HBM4, based on its 10nm-class 6th-generation (1c) DRAM, is expected to be delayed until next year. The company had initially aimed for mass production in the second half of this year, but is now taking a more cautious approach, setting the target for Production Readiness Approval (PRA) in the fourth quarter. However, the encouraging news is that the redesign process for the 1c DRAM shows signs of improved performance and yield. A source familiar with the matter stated, "The company has internally set the goal for HBM4 12-stack PRA in Q4. This means mass production is now targeting next year, not this year." The source added, "Based on small-scale sample tests conducted at the R&D fab, the 1c DRAM yield recorded 65%, which is a hopeful situation." Typically, memory companies first develop DRAM dies for computing, then expand to derivative products like mobile or HBM. The 1c DRAM that recently passed PRA earlier this month is also not an HBM-specific core die. The XX% yield is analyzed to be from the 1c DRAM sample production test that the company had planned for this month. In actual mass production, the yield may be adjusted, so it is expected that the company will aim for a higher sample product yield before entering mass production. For now, Samsung Electronics is expected to deliver mass production samples of HBM4, based on these sample production tests, to key customers within this quarter. During its Q1 earnings conference call, Samsung Electronics repeatedly emphasized, "For HBM4, we are developing it with the goal of mass production in the second half, as planned, to align with customer project schedules." However, it is now highly likely that the HBM4 mass production timeline will be pushed to next year. The general consensus is that even if HBM PRA is quickly approved in Q4, it will still take some time to reach the mass production stage. Even SK Hynix, which currently holds a dominant position in the HBM market, has indicated the possibility of mass producing HBM4 next year. While the mass production timeline is being pushed back, the positive direction of the 1c DRAM redesign process is considered encouraging. Samsung Electronics previously conducted the 1c DRAM redesign in two main ways. One involved newly developing 1c through similar designs of 1a and 1b, while the other involved redesigning both 1a and 1b to implement a completely new 1c. It is reported that the latter approach yielded superior results when the company attempted both simultaneously. The analysis suggests that by redesigning both 1a and 1b, the chip size of the 1c DRAM was increased, leading to an improvement in yield. However, the expanded scope of redesign naturally led to increased cost burden. Another industry source commented, "The mere act of redesigning both 1a and 1b incurred enormous costs, and the number of chips that can be produced also decreased, leading to reduced margins." The key challenge for Samsung Electronics is whether it can maintain the yield achieved in the sample test phase during mass production. Samsung Electronics still has additional tests remaining to apply this developed 1c DRAM to the mass production stage. The aforementioned source stated, "Interpreting PRA passage as securing victory is an overstatement. Yield issues arise when moving into the mass production stage." They added, "It will still take time to stably secure yield alongside increasing HBM4 production."  XXXXX engagements  **Related Topics** [dram](/topic/dram) [$ssnlf](/topic/$ssnlf) [$005930ks](/topic/$005930ks) [samsung](/topic/samsung) [Post Link](https://x.com/Jukanlosreve/status/1948166001083961430)
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]
Jukan @Jukanlosreve on x 22.6K followers
Created: 2025-07-23 23:39:04 UTC
Report: Samsung Electronics' HBM4 Mass Production Expected 'Next Year'... Sample Test Yield at XX%
Samsung Electronics' mass production of 12-stack HBM4, based on its 10nm-class 6th-generation (1c) DRAM, is expected to be delayed until next year. The company had initially aimed for mass production in the second half of this year, but is now taking a more cautious approach, setting the target for Production Readiness Approval (PRA) in the fourth quarter. However, the encouraging news is that the redesign process for the 1c DRAM shows signs of improved performance and yield.
A source familiar with the matter stated, "The company has internally set the goal for HBM4 12-stack PRA in Q4. This means mass production is now targeting next year, not this year." The source added, "Based on small-scale sample tests conducted at the R&D fab, the 1c DRAM yield recorded 65%, which is a hopeful situation."
Typically, memory companies first develop DRAM dies for computing, then expand to derivative products like mobile or HBM. The 1c DRAM that recently passed PRA earlier this month is also not an HBM-specific core die. The XX% yield is analyzed to be from the 1c DRAM sample production test that the company had planned for this month. In actual mass production, the yield may be adjusted, so it is expected that the company will aim for a higher sample product yield before entering mass production. For now, Samsung Electronics is expected to deliver mass production samples of HBM4, based on these sample production tests, to key customers within this quarter.
During its Q1 earnings conference call, Samsung Electronics repeatedly emphasized, "For HBM4, we are developing it with the goal of mass production in the second half, as planned, to align with customer project schedules." However, it is now highly likely that the HBM4 mass production timeline will be pushed to next year. The general consensus is that even if HBM PRA is quickly approved in Q4, it will still take some time to reach the mass production stage. Even SK Hynix, which currently holds a dominant position in the HBM market, has indicated the possibility of mass producing HBM4 next year.
While the mass production timeline is being pushed back, the positive direction of the 1c DRAM redesign process is considered encouraging. Samsung Electronics previously conducted the 1c DRAM redesign in two main ways. One involved newly developing 1c through similar designs of 1a and 1b, while the other involved redesigning both 1a and 1b to implement a completely new 1c.
It is reported that the latter approach yielded superior results when the company attempted both simultaneously. The analysis suggests that by redesigning both 1a and 1b, the chip size of the 1c DRAM was increased, leading to an improvement in yield. However, the expanded scope of redesign naturally led to increased cost burden. Another industry source commented, "The mere act of redesigning both 1a and 1b incurred enormous costs, and the number of chips that can be produced also decreased, leading to reduced margins."
The key challenge for Samsung Electronics is whether it can maintain the yield achieved in the sample test phase during mass production. Samsung Electronics still has additional tests remaining to apply this developed 1c DRAM to the mass production stage. The aforementioned source stated, "Interpreting PRA passage as securing victory is an overstatement. Yield issues arise when moving into the mass production stage." They added, "It will still take time to stably secure yield alongside increasing HBM4 production."
XXXXX engagements
