Skip Navigation

Rekall Incorporated

@ Rekall_Incorporated @piefed.social

Posts
339
Comments
284
Joined
9 mo. ago

Rekall is a company that provides memory implants of vacations, where a client can take a memory trip to a certain planet and be whoever they desire.

  • AMD's legacy Ryzen 7 5800X3D chips now sell for up to $800, more than a new 9800X3D — AM4 chip costs twice as much as MSRP, as enthusiasts flock to old DDR4 memory

    Jump

  • Yeah, I would just not risk doing a downgrade at this point. Where I live prices are elevated in general (import duties, less common and/or expensive SKUs tend to pieced even bigger) and things are only going to get worse.

    With RAM specifically it can even be difficult to find certain high performance variants (this was back when DDR4 was being mass produced) at all.

  • Exynos 2600 is fundamentally different than Samsung's previous in-house chips

    Jump

  • We'll see how well it performs with non-Samsung modems. I have my doubts, but I would only be happy for more competition for Qualcomm and Mediatek.

  • Arm stock declines massively as Qualcomm acquires RISC-V designer Ventana

    Jump

  • Thankfully Qualcomm can't "buy out RISC-V". I think the current size of the ecosystem and competitive dynamics are such they can't kill RISC-V even in an in-direct manner.

    That being said, I haven't really seen anything all too impressive on the RISC-V front.

    I would be willing to switch my SBC to RISC-V as I only run open source applications on Linux and I don't even need transcoding.

  • AMD's legacy Ryzen 7 5800X3D chips now sell for up to $800, more than a new 9800X3D — AM4 chip costs twice as much as MSRP, as enthusiasts flock to old DDR4 memory

    Jump

  • I meant dealing with the hassle of downgrading and then upgrading the RAM. Might be easier to stick with what you have and not risk it with current shortages.

    For sure, I am taking in a general manner.

  • Moore Threads unveils next-gen gaming GPU with 15x performance and 50x ray tracing improvement — AI GPU with claimed performance between Hopper and Blackwell also in the works

    Jump

  • It's not going to be well supported unless the Nvidia and AMD are required to make their pre-2020 driver code public domain. Which is not going to happen.

  • India unveils a homegrown dual-core 1GHz RISC-V processor

    Jump

  • Agreed, but then government should provide some sort of guarantee that it will be available at least for another ~10 years. Or do so for the next release.

  • AMD's legacy Ryzen 7 5800X3D chips now sell for up to $800, more than a new 9800X3D — AM4 chip costs twice as much as MSRP, as enthusiasts flock to old DDR4 memory

    Jump

  • Edit: now he’s considering if he just wants to sell it and go without gaming for a while.

    IMO, not worth it unless the RAM is not in use.

  • Linux 6.19 Lands Fix For Seagate Barracuda HDD Taking Down The SATA Bus

    Jump

  • Pretty brutal bug even though it was tied to a specific model.

  • AMD's legacy Ryzen 7 5800X3D chips now sell for up to $800, more than a new 9800X3D — AM4 chip costs twice as much as MSRP, as enthusiasts flock to old DDR4 memory

    Jump

  • It was, I believe the 5700X3D is also not in production.

  • Tired of sky-high memory prices? Buckle up, we're in this for the long haul - We haven't even hit the peak, TechInsights tells El Reg

    Jump

  • An interesting chart, but difficult to interpret in more depth. The article suggests this refers to price changes not "market growth", but then what is the unit per price value?

    According to TechInsights, memory pricing was already on the rise in 2024, growing 88 percent from a rather steep valley the year prior. Based on previous DRAM booms, one might expect it to grow at a slower pace in 2025, before contracting in 2026 or 2027.

  • AMD's legacy Ryzen 7 5800X3D chips now sell for up to $800, more than a new 9800X3D — AM4 chip costs twice as much as MSRP, as enthusiasts flock to old DDR4 memory

    Jump

  • From my perspective, a high end (at the time of purchase) desktop should still be usable in 10 years.

    In 2023 I was working on an old laptop from 2014 with 760M and an i7-4702MQ, it was not the best (although I added an SSD and upgraded to 16GB RAM), but it did OK and could play older games just fine.

  • AMD's legacy Ryzen 7 5800X3D chips now sell for up to $800, more than a new 9800X3D — AM4 chip costs twice as much as MSRP, as enthusiasts flock to old DDR4 memory

    Jump

  • I really wish they did another 5800X3D run (or even a 5800X3D XT). I missed the opportunity to buy it and the grey market prices are stupid.

    I have a pretty solid AM4 rig; 3080 GPU (which is enough for my needs), x2 32 GB 16 CL 3600 memory. Tons of storage for HDD, PCIe 4 and PCIe 3 for different use cases, a solid cooling system.

    I was planning to upgrade to AM5 next year, but I don't think it's happening. To be honest, the only way I am limited is the CPU.

  • Nvidia reportedly plans 30-40% cut in GeForce GPU production in early 2026

    Jump

  • I don't think Chinese GPU vendors will be any better than Nvidia or AMD. That being said, Chinese GPU companies taking say 25% of the (global, B2C) market will probably force Nvidia and AMD to change their tone.

  • Taiwan considers TSMC export ban that would prevent manufacturing its newest chip nodes in U.S. — limit exports to two generations behind leading-edge nodes, could slow down U.S. expansion

    Jump

  • One thing that is undeniably true is that the Taiwanese cannot trust the Americans. A two generation gap is very reasonable on the part of Taiwan.

  • Radiant bags $300M-plus to commercialize its microreactors

    Jump

  • Unlike other SMR designs we've discussed over the past year, Radiant's Kaleidos reactors are tiny, capable of producing about a megawatt of power each. However, the actual mechanism for energy generation isn't too far off from the designs we've seen from X-Energy, which use TRISO fuel pellets and helium gas as a coolant.

    I am not expert on DC power provision strategy, but one megawatt seems like a tiny amount.

    However, the jury is still out on whether SMRs will ever be cost effective. In September, analysts at the Centre for Net Zero (CNZ) estimated it would cost 43 percent less to power a 120 MW data facility with renewables and a small amount of gas-generated energy, compared to using SMRs

    Then what's the point? Even if economies of scale bring down production costs, there are still going to be unique costs with operating SMRs.

    Are they planning to deploy them without any safety measures? Literally just a trailer module that any well-connected local goon can deploy anywhere?

  • Some of my Anno 117 Cities [2025-12-17]

    Jump

  • US tech firm Palantir extends deal with French intelligence agency

    Jump

  • I have mixed views on this.

    On one hand I agree with you, especially when it comes to dealing with Palantir or really any company that can be influenced by the US, but on the other hand there are legitimate uses for such technologies in the sphere of national security and even public security.

    I would argue it's the citizens' responsibility to make sure that the usage of such technologies is done in a framework of checks and balances (i.e. in a responsible manner).

    I don't believe in rhetoric about "the state infinitely expands surveillance capabilities". The state is a reflection of the voters and there is no laws of physics or chemistry that guarantees such expansion via Brownian motion or what have you. If you do have institutions going overboard (be it the state or corporations), the root cause are the citizens (examples like NK or Eritrea notwithstanding).

  • US tech firm Palantir extends deal with French intelligence agency

    Jump

  • Incredible that they're still contracting an American company (one that is known to be controlled by particularly corrupt oligarchs) instead of building out their own capabilities.

  • Nanoimprint over EUV Lithography: Japan Aims to Cut Costs for 1.4nm Chip Production

    Jump

  • In the fall of 2024, the first samples of the equipment were sent to Intel for evaluation. Later, Samsung, TSMC, and other major market players also showed interest in the technology. DNP plans to start mass production of the necessary materials for 1.4nm chips in 2027. However, the established industry, entirely built around photolithography, could slow the adoption of the new technology. Switching to it would require manufacturers to significantly retool their existing production lines.

    Sounds like ecosystem/industry inertia will limit the adoption. Perhaps Japan's Rapidus will try and leverage this tech as sort of high risk / high reward strategy to compete against TSMC.