Matt
95 posts
※パクリ禁止
直近のメモリ価格の材料一覧（■下げの主因）
■Semianalysys がRubin NVL72 SOCAMM DRAMを
ラックあたり約55TBから約28TBに削減・192GBから96GBに移行報道あり
・コスト$7.6Mから$6.8Mに低下
＝＝この報道がどれだけ正確か検証
①SOCAMMはLPDDR5Xであり大きな収益源「HBM4は依然として8スタック288GBで変更なし」
②数量変化の例 192GB→半減 96GB→6倍 64GB→50%増
③コスト低減によりラック普及を早め、出荷積極拡大のための「仕様調整」でしかない
④SoCAMM2 LPDDRの総需要・TAMは拡大期待あり
以前の予測に対し10〜20%増加するとの指摘も
結論
【メモリ「需要の総量が減る話ではない」可能性大】
＝＝
：余談
この件そのものについてではないが、質疑にも言及あり
Micron CEO「DRAM不足により需要側は計画調整し、少ないDRAMに適応する方法を考える必要あり」
NVDAに限らず、メーカー側が供給不足をあらゆる工夫をするのは当たり前
例：「ゲーム機の搭載メモリを減らす」とか
【それ自体が需要緩和には全くつながらない】
＝＝他メモリ 弱気材料まとめ
■CEOが株売り（量は8万株のみ）
→「特別な理由のない内部者売りは計画的なものもありほぼ無意味」は常識
■BNPパリバ
DRAMとNANDのASPが26年半ばにピークを迎え来年初頭から四半期ごとの減少に転じると予想
当初の27年半ばという予測よりもかなり早い。
理由：中国CXMTやYMTCによる積極的な生産能力拡大と価格高騰による他産業のメモリー消費減少が組み合わさり、今年の世界的なスマホ出荷14％減も理由の一つ
今後1～2年以内の過剰供給と利益率悪化への懸念が高まりそれがすでにこのセクターに事前の圧力をかけ始めていると指摘
→主要メーカーは競合の動向や市況予測は最も把握できる立場であ り、それが本当ならなぜ「兆円単位の先払い」や「3－5年の長期契約」を結ぶのか。まったく整合性が取れないので個人的にスルー
ただし・・「AI向けを除くコモディティNAND」市場が急速に需要を満たす予想は事実
YMTCのシェアは恐らく来年にかけて急増し、安価なNANDの価格低下にはつながるというのが大方の予想
Kioxiaにとっての直接的な課題だが、その頃までに汎用品比率を40%に下げ（DC比率60%）対応するという話
あくまで例だが、その40%のうち半数程度の20％の利益が半減しても影響は10％で極めて軽微
＝＝強気材料
■SKコメント
「Vera Rubinに使用されるLPDDRメモリの需要が強く伸びている」
「新しいNAND容量を追加する計画はありません」
→Kioxiaに相対的に＋ 今日のメモリ下落の中で上昇寄与した可能性あり？
＝＝ここで豆知識
MUってアホみたいにすぐ下がるけどナニコレ！？
↓
【ハイベータ株の‐10～30％は数か月に一度の通常営業】
過去のドローダウンの例。
■3月 GoogleのKVC圧縮TurboQuant
450ドル→315ドル = -30%
■5月 イラン戦争後の調整5月
818ドル→652ドル= -20％
■6月 今回地合い＋Rubin報道下げ
1080ドル→943= -13%
統計的に振り返ると、
個人的にもその前提で、資金管理を上手くしたいと思います。
めろーね@melone3710
MUの主要な下落理由はこのRubinのメモリ減らす見込みって話だと思いますが、 （あとサイクル指摘の低いレーティングのレポート 自分がまとめた中にあって既出なのよね・・ 短期的な反応は本当にわからない・・ 短期は方向感確認で雇用・金曜OPもあってロット落として個別も利益死守でノーポジだけど長期はそのままでいいかなと・・ >>リポスト内 【メモリ価格高騰によりNVDAはVeraRubinのLPDDR量削減も議論中との情報もあり】
日本語
@dylan522p see what's the truth？actually just optimized the rate for different varieties，overall the quantity is increased！
English
I love that people reposting what we said miss most of what the note says.
Happens all the time.
Jukan @COMPUTEX@jukan05
SemiAnalysis: NVIDIA is reducing the SOCAMM capacity for Rubin. via @TMTBreakout
Taipei City, Taiwan 🇹🇼 English
$MU $DRAM Der Markt interpretiert das falsch, das ist nur noise. $NVDA optimiert einfach die specs bzw. Kosten da die Preise so gestiegen sind und um Lieferverzögerungen zu verhindern.
SOCAMM 192GB Bestellmenge halbiert sich zwar, aber 96GB und 64GB Bestellmengen erhöhen sich. Insgesamt erhöht sich das Shipment Volumen und der Bedarf an SOCAMM steigt sogar.
Und HBM4 welcher den Umsatz antreibt bleibt unverändert.
Jukan @COMPUTEX@jukan05
SemiAnalysis: NVIDIA is reducing the SOCAMM capacity for Rubin. via @TMTBreakout
Deutsch
So SemiAnalysis dropped a headline that looks bearish for memory, and everyone is misreading it. $MU
The claim: NVDA is cutting Rubin NVL72 SOCAMM DRAM from ~55TB to ~28TB per rack. Most Rubin systems move to 96GB modules instead of 192GB. Rack cost drops from $7.6M to $6.8M, TCO from $4.16 to $3.90 per GPU-hour.
Surface read: they are halving the memory. Sell memory.
Here is the part most people get wrong.
SOCAMM is LPDDR5X. It is not HBM. The thing they trimmed is the commodity DRAM bolted to the Vera CPU.
The HBM4 on the actual Rubin GPU is untouched, still 8 stacks, 288GB per GPU. The high-margin link in the chain did not move an inch.
So why cut at all?
Because memory is the current bottleneck and NVDA knows it.
SemiAnalysis has SOCAMM contract pricing near $8/GB in 1Q26, a sharp step-up driven by the LPDDR5X surge and broad DRAM tightness.
When your input is scarce and repricing higher every quarter, you ration it. You spec it down. You manage TCO.
That is not demand weakness. That is a buyer flinching at how tight supply is.
→ NVDA still calls SOCAMM a $300B TAM. They are managing specs inside a shortage, not walking away.
→ The cut is on the cheap DRAM. The scarce, sold-out, margin-rich part is HBM, and that allocation is locked.
→ MU confirmed its HBM4 36G 12H in high-volume production for Vera Rubin. At GTC Taipei NVDA named all three suppliers. SK Hynix takes the largest share, Samsung next, MU the smallest slice. But MU is in the build.
The frame I keep coming back to: the bottleneck moved from chips to memory, and the durable one is power.
This SOCAMM story is NVDA actively managing the memory link. You do not manage what is abundant.
The risks remain:
→ $MU is the smallest of the three on Rubin HBM4. Share, not access, is the open question.
→ If LPDDR pricing is what NVDA is really pushing back on, that pressures commodity-DRAM margin even as HBM holds.
A headline that screams bearish is NVDA confirming memory is the choke point.
They cut the cheap stuff and protected the expensive stuff.
That is the whole tell.
English
@cherryPayment K线的话是这样，今天下去也是回补缺口的，补完就上去了。 之前群里说这个票很吃技术。我观察了几天确实。 但还是容易受消息面波动，这时候技术就失效了
中文
Wow… new extremely transformative news got released today.
Making a certain photonics company:
The effective upstream laser chokepoint for $NVDA NVLink fusion CPO ecosystem.
With their lasers now in Nvidia’s optical infrastructure supply chains.
Can anyone guess the name?
English
@Ren_aramb Damn, just stop spreading the good news, you realize that it will limit the chance to load up more at a lower price?!
Bro, I am waiting for my monthly trust fund pay check.
English
Bullish $SIVE
Ayar Labs just joined the NVIDIA NVLink Fusion ecosystem.
CPO is now foundational infrastructure for AI factories.
Think of CPO like replacing the copper wires inside a server with fiber optic cables soldered directly onto the chip. Less heat. Less power. More bandwidth. The interconnect problem is solved at the die level.
Ayar Labs needs an External Light Source to feed those optical engines. That laser has to be continuous wave, high coherence, built in InP. That is exactly what $SIVE makes.
Sivers and Ayar Labs co-founded the CW-WDM MSA together. They did not just watch this ecosystem get built – they helped write the spec.
When Nvidia blesses a technology at the NVLink Fusion level, the supply chain behind it re-rates.
Yeah, we are not nearly done with re rating the CPO supply chain.
Massively bullish CPO.
Long $SIVE.
Ayar Labs@AyarLabs
Today, @AyarLabs announced it has joined the @nvidia NVLink Fusion ecosystem, introducing co-packaged optics as a foundational building block for hyperscalers and system innovators deploying heterogeneous compute in NVIDIA AI factories. Press Release: bit.ly/4oa8epa
English
Ansem on $BB again
这是建仓了吧，感谢同车
Let’s ride.
猎手killer🪙@memekiller365
Welcomet to Blackberry , Ansem . $BB
中文
$BB (June 2, 2026-daily chart)
If you want to succeed in the stock market, learn to invest — not just trade in and out. When you’re in the right stocks, patience is key.
Congratulations to all $BB shareholders, especially my community who loaded up between $4.50–$5.50. Special thanks to LEE @sheslee for sharing her $BB thesis and — massive congratulations on your huge gains with over 350,000 shares!
@sheslee
Danny cheng@dannycheng2022
First of all, I want to apologize for sharing my buy orders for $BB between $4.5 and $5.5. Instead of rising, the stock dropped into the $3 range. That said, as I’ve always told my community, I’m not a trader. For a company I believe in, I chose to hold rather than cut losses. Today, my patience has finally paid off. Congratulations to all $BB shareholders! A special shoutout to my community members who supported me through the tough times and stayed in the stock with me. A huge congratulations to one of my loyal subscribers, Stephen, who went big on leaps and turned it into an impressive US$528,000 profit. Thank you for trusting me and supporting me since February 2024.
English
Breaking news! 💥
Sivers $SIVE & GlobalFoundries $GFS Advance AI Data Center Optical Solutions
Tue, Jun 02, 2026 07:00 CET
Sivers’ laser arrays to support GlobalFoundries’ silicon photonics platform and SCALE™ optical engine solutions targeting a $25B Pluggable Optics market by 2030
Kista, Sweden – June 2, 2026 – Sivers Semiconductors AB (STO:SIVE), a global leader in photonics and wireless technologies, today announced a strategic collaboration with GlobalFoundries (Nasdaq: GFS) (GF), to develop advanced silicon photonics solutions for the high-growth AI infrastructure market.
Sivers Semiconductors’ laser arrays will be integrated into reference designs built on GF’s silicon photonics platform. The collaboration supports a range of optical connectivity architectures, including co-packaged optics (CPO), linear pluggable optics (LPO), and other emerging data center interconnect solutions. Sivers’ laser arrays will also be available in GF’s Silicon Photonics Co-packaged Advanced Light Engine (SCALE™) platform for next-generation optical sub-assemblies and light engine architectures. GF’s SCALE CPO solution combines integrated photonic devices, coarse and dense wavelength-division multiplexing (CWDM, DWDM) and advanced packaging enablement to improve bandwidth density and system scalability.
news.cision.com/sivers-semicon…
English
We're excited to announce a strategic collaboration with @GlobalFoundries to develop advanced silicon photonics solutions for the rapidly growing AI infrastructure market.
Our laser arrays will support next-generation optical connectivity architectures, including CPO, LPO, and other emerging data center interconnect solutions, while complementing GlobalFoundries' silicon photonics platform and SCALE™ optical engine offerings targeting a projected $25 billion pluggable optics market by 2030.
For full details, visit: sivers-semiconductors.com/press/sivers-g…
English
English
@asianinvestors i saw this post before.and now may he crying in the toilet
English
Apparently he's some famous Swedish investor on X
But he says $SIVE only worth 3 SEK. I'm sure he doesn't believe that because that's ridiculous. But when he exaggerates, people listen. I'm sure he's had some great calls, but right now, as far as I'm concerned he's wrong.
Doesn't matter if in 10 years it goes down to 20SEK. Being an early bear is still wrong if the price moves against you so significantly
English
