Looks like you're in the United States. Switch to the United States site for products available in your region.
Post
taishiro
Mostra originale
The HBM trade is becoming a capacity allocation story, not just an AI demand story.
When memory inventory gets this tight, chipmakers face a choice: prioritize high margin HBM or supply conventional DRAM. That competition can push memory prices higher, but it also makes supply normalization the biggest risk to the trade.
SK Hynix has the advantage today, but Samsung's progress shows that leadership is not guaranteed.
#SamsungQ3ProfitOutlook
ARRIWYKS
#SKhynixQ3HBM4Earnings
#Hynix Cloud +AI HBM4 Cloud = Collo di bottiglia della memoria ora, non GPU
Mercato HBM 38 miliardi di $→58 miliardi di $ +52% 26E, 85 miliardi di $ 27E. SK = 62% quota HBM, *∼70% di NVDA Vera Rubin HBM4* (TrendForce) vs 50% stimato. 16 stack = 576GB/stack vs HBM3E 8Gb → 2x BW.
Perché ottimisti: Inventario <10 giorni (primo mai registrato). Wafer 150-200k/mese 3-4x Micron, 50% → HBM4 = DDR5 compresso → Q4 DRAM +10-15% NAND +15-20%. JPM vede 20%
*Rischi:* Samsung 8,7 miliardi $ HBM Q3 > SK 7,6 miliardi $ per la prima volta*
*OKX play: Long leva HBM, short PC DDR.*
Disclaimer: i contenuti di OKX Orbit sono forniti solo a scopo informativo. Scopri di più
