Hacker News: Deathmax

New comment by Deathmax in "Veracrypt project update"

Deathmax — Wed, 08 Apr 2026 16:16:28 +0000

Microsoft has been taking steps to mitigate the leaked code signing certificate problem.

On the driver side of things, new versions of Windows no longer trust the cross-signed certs, so you must submit your driver to Microsoft to validate and sign, so no private key to go missing. https://techcommunity.microsoft.com/blog/windows-itpro-blog/...

On the regular Authenticode side of things, the new CA/B Forum rules have prohibited storing new private keys outside of hardware modules for a while now, so eventually you won't be able to find a leaked private key for code signing that would still be valid.

New comment by Deathmax in "VPN location claims don't match real traffic exits"

Deathmax — Sat, 13 Dec 2025 22:14:11 +0000

NordVPN calls out when a location is virtual, so unless ipinfo is claiming they have virtual locations that are not labelled as such, they are at least transparent about it. They did document the physical server locations of their virtual locations at launch, but I'm not sure if there's a live doc for new locations. https://nordvpn.com/blog/new-nordvpn-virtual-servers/

New comment by Deathmax in "I got an Nvidia GH200 server for €7.5k on Reddit and converted it to a desktop"

Deathmax — Thu, 11 Dec 2025 13:33:40 +0000

The author was running a quantised version of GLM 4.5 _Air_, not the full fat version. API pricing for that is closer to $0.2/$1.1 at the top end from z.ai themselves, half the price from Novita/SiliconFlow.

New comment by Deathmax in "Nano Banana Pro"

Deathmax — Thu, 20 Nov 2025 16:38:28 +0000

You sort of can on Android, but it's a few steps:

1. Trigger Circle to Search with long holding the home button/bar

2. Select the image

3. Navigate to About this image on the Google search top bar all the way to the right - check if it says "Made by Google AI" - which means it detected the SynthID watermark.

New comment by Deathmax in "Google Antigravity"

Deathmax — Wed, 19 Nov 2025 11:13:24 +0000

Depending on which shared GCP project you get assigned to, mine had a global 300 million tokens per minute quota that was being hit regularly.

New comment by Deathmax in "Gemini 3 Pro Model Card [pdf]"

Deathmax — Tue, 18 Nov 2025 14:50:13 +0000

Bad news then, they've bumped 3.0 Pro pricing to $2/$12 ($4/$18 at long context).

New comment by Deathmax in "Steam Machine"

Deathmax — Sat, 15 Nov 2025 12:56:20 +0000

So the Bigscreen Beyond 1 and 2? They've optimised entirely for weight at ~100 grams.

New comment by Deathmax in "Tailscale Peer Relays"

Deathmax — Wed, 29 Oct 2025 18:02:08 +0000

I can finally tear down my custom DERP server that I was using to get higher throughput between two NAT'd clients.

New comment by Deathmax in "Tell HN: OpenAI now requires ID verification and won't refund API credits"

Deathmax — Sat, 25 Oct 2025 14:32:20 +0000

Not anymore, especially after other routers like Vercel's AI Gateway and proxies from LLM providers like Fal, DeepInfra, and AtlasCloud didn't get the memo of enforcing BYOK for ID verification required models after GPT-5's release.

New comment by Deathmax in "Apple loses UK App Store monopoly case, penalty might near $2B"

Deathmax — Thu, 23 Oct 2025 23:55:29 +0000

As if the App Store had any sort of those guarantees. I know of people have been scammed via WebView wrappers that purported to be some benign app to pass app store review, which were then pointed at fake exchange websites afterwards. GitLab which was hosting their C&C mechanism took action faster than Apple or Google did to take down multiple scam apps across multiple different developer identities, but the scammers spun up new apps the next day.

New comment by Deathmax in "Is it possible to allow sideloading and keep users safe?"

Deathmax — Sun, 31 Aug 2025 22:04:45 +0000

Victims would almost certainly have transferred funds to money mules, who would have then immediately broken digital audit trails to the ultimate destination by withdrawing as cash before passing it around.

New comment by Deathmax in "Secure Boot, TPM and Anti-Cheat Engines"

Deathmax — Tue, 19 Aug 2025 12:21:29 +0000

Riot documents the need to have IOMMU support enabled for Vanguard: https://support-valorant.riotgames.com/hc/en-us/articles/222...

New comment by Deathmax in "Batch Mode in the Gemini API: Process More for Less"

Deathmax — Fri, 11 Jul 2025 14:39:37 +0000

Vertex's offering of Gemini very much does implicit caching, and has always been the case [1]. The recent addition of applying implicit cache hit discounts also works on Vertex, as long as you don't use the `global` endpoint and hit one of the regional endpoints.

[1]: http://web.archive.org/web/20240517173258/https://cloud.goog..., "By default Google caches a customer's inputs and outputs for Gemini models to accelerate responses to subsequent prompts from the customer. Cached contents are stored for up to 24 hours."

New comment by Deathmax in "Show HN: TokenDagger – A tokenizer faster than OpenAI's Tiktoken"

Deathmax — Mon, 30 Jun 2025 14:56:55 +0000

Gemini uses SentencePiece [1], and the proprietary Gemini models share the same tokenizer vocabulary as Gemma [2, 3, 4].

Out of the large proprietary western AI labs (OpenAI, Anthropic, Google), only Anthropic with Claude 3 and newer lack local tokenizers.

[1] https://github.com/google/sentencepiece

[2] https://github.com/googleapis/python-aiplatform/blob/main/ve...

[3] https://storage.googleapis.com/deepmind-media/gemma/gemma-2-...: "We inherit from the large Gemini vocabulary (256k entries)."

[4] https://storage.googleapis.com/deepmind-media/gemma/Gemma3Re...: "We use the same tokenizer as Gemini 2.0."

New comment by Deathmax in "I made my VM think it has a CPU fan"

Deathmax — Sun, 29 Jun 2025 20:26:09 +0000

It's a change to the CA rules that was passed in https://cabforum.org/2022/04/06/ballot-csc-13-update-to-subs... to align OV certificate requirements with the EV ones (that enforces the use of HSMs/hardware tokens/etc) that was meant to go into effect for new certificates issued after November 2022, but was delayed and eventually implemented on June 1 2023.

New comment by Deathmax in "Show HN: Ts-SSH – SSH over Tailscale without running the daemon"

Deathmax — Fri, 20 Jun 2025 16:37:33 +0000

Since April 2023 they support custom OIDC providers[1], and as of April 2024 that was extended to the free plan as well[2], so you can bring your own auth.

[1]: https://tailscale.com/kb/1240/sso-custom-oidc

[2]: https://tailscale.com/blog/sso-tax-cut

New comment by Deathmax in "MiniMax-M1 open-weight, large-scale hybrid-attention reasoning model"

Deathmax — Wed, 18 Jun 2025 11:46:31 +0000

https://www.minimaxi.com is their website for the Chinese parent company 上海稀宇科技有限公司, https://minimax.io is their international website for the Singapore based company Nanonoble Pte Ltd that handles operations outside of China.

New comment by Deathmax in "OpenAI dropped the price of o3 by 80%"

Deathmax — Tue, 10 Jun 2025 19:38:23 +0000

I did read it, and I even went to their eval repo.

> At the time of writing, there are two major versions available for GPT-4 and GPT-3.5 through OpenAI’s API, one snapshotted in March 2023 and another in June 2023.

openaichat/gpt-3.5-turbo-0301 vs openaichat/gpt-3.5-turbo-0613, openaichat/gpt-4-0314 vs openaichat/gpt-4-0613. Two _distinct_ versions of the model, and not the _same_ model over time like how people like to complain that a model gets "nerfed" over time.

New comment by Deathmax in "OpenAI dropped the price of o3 by 80%"

Deathmax — Tue, 10 Jun 2025 19:22:36 +0000

Your linked article is specifically comparing two different versioned snapshots of a model and not comparing the same model across time.

You've also made the mistake of conflating what's served via API platforms which are meant to be stable, and frontends which have no stability guarantees, and are very much iterated on in terms of the underlying model and system prompts. The GPT-4o sycophancy debacle was only on the specific model that's served via the ChatGPT frontend and never impacted the stable snapshots on the API.

I have never seen any sort of compelling evidence that any of the large labs tinkers with their stable, versioned model releases that are served via their API platforms.

New comment by Deathmax in "Devstral"

Deathmax — Wed, 21 May 2025 21:13:09 +0000

Also known as Claude 3.5 Sonnet V2 on AWS Bedrock and GCP Vertex AI