Fernandes

@BMKing23 this game is a mess

There’s something weird going on with DLSS in #CrimsonDesert. For some reason, DLAA looks softer than DLSS Quality, Balanced, or even Performance. This affects both DLSS 4 and 4.5 presets.

@clips_montage @EPlCs @ArtificialAnlys @MistralAI Europe literally doesn't have anything else

@EPlCs @ArtificialAnlys @MistralAI nah just french model europe doesnt wanna claim this🥀

Mistral has released Mistral Small 4, an open weights model with hybrid reasoning and image input, scoring 27 on the Artificial Analysis Intelligence Index @MistralAI's Small 4 is a 119B mixture-of-experts model with 6.5B active parameters per token, supporting both reasoning and non-reasoning modes. In reasoning mode, Mistral Small 4 scores 27 on the Artificial Analysis Intelligence Index, a 12-point improvement from Small 3.2 (15) and now among the most intelligent models Mistral has released, surpassing Mistral Large 3 (23) and matching the proprietary Magistral Medium 1.2 (27). However, it lags open weights peers with similar total parameter counts such as gpt-oss-120B (high, 33), NVIDIA Nemotron 3 Super 120B A12B (Reasoning, 36), and Qwen3.5 122B A10B (Reasoning, 42). Key takeaways: ➤ Reasoning and non-reasoning modes in a single model: Mistral Small 4 supports configurable hybrid reasoning with reasoning and non-reasoning modes, rather than the separate reasoning variants Mistral has released previously with their Magistral models. In reasoning mode, the model scores 27 on the Artificial Analysis Intelligence Index. In non-reasoning mode, the model scores 19, a 4-point improvement from its predecessor Mistral Small 3.2 (15) ➤ More token efficient than peers of similar size: At ~52M output tokens, Mistral Small 4 (Reasoning) uses fewer tokens to run the Artificial Analysis Intelligence Index compared to reasoning models such as gpt-oss-120B (high, ~78M), NVIDIA Nemotron 3 Super 120B A12B (Reasoning, ~110M), and Qwen3.5 122B A10B (Reasoning, ~91M). In non-reasoning mode, the model uses ~4M output tokens ➤ Native support for image input: Mistral Small 4 is a multimodal model, accepting image input as well as text. On our multimodal evaluation, MMMU-Pro, Mistral Small 4 (Reasoning) scores 57%, ahead of Mistral Large 3 (56%) but behind Qwen3.5 122B A10B (Reasoning, 75%). Neither gpt-oss-120B nor NVIDIA Nemotron 3 Super 120B A12B support image input. All models support text output only ➤ Improvement in real-world agentic tasks: Mistral Small 4 scores an Elo of 871 on GDPval-AA, our evaluation based on OpenAI's GDPval dataset that tests models on real-world tasks across 44 occupations and 9 major industries, with models producing deliverables such as documents, spreadsheets, and diagrams in an agentic loop. This is more than double the Elo of Small 3.2 (339) and close to Mistral Large 3 (880), but behind gpt-oss-120B (high, 962), NVIDIA Nemotron 3 Super 120B A12B (Reasoning, 1021), and Qwen3.5 122B A10B (Reasoning, 1130) ➤ Lower hallucination rate than peer models of similar size: Mistral Small 4 scores -30 on AA-Omniscience, our evaluation of knowledge reliability and hallucination, where scores range from -100 to 100 (higher is better) and a negative score indicates more incorrect than correct answers. Mistral Small 4 scores ahead of gpt-oss-120B (high, -50), Qwen3.5 122B A10B (Reasoning, -40), and NVIDIA Nemotron 3 Super 120B A12B (Reasoning, -42) Key model details: ➤ Context window: 256K tokens (up from 128K on Small 3.2) ➤ Pricing: $0.15/$0.6 per 1M input/output tokens ➤ Availability: Mistral first-party API only. At native FP8 precision, Mistral Small 4's 119B parameters require ~119GB to self-host the weights (more than the 80GB of HBM3 memory on a single NVIDIA H100) ➤ Modality: Image and text input with text output only ➤ Licensing: Apache 2.0 license

@RyukTheShnigami @BMKing23 Frame-gen was causing me a lot of stuttering, turned it off and it is now mostly silky smooth.

@BMKing23 Dude this game seems to having issues with people running on amd X3D cpus. Including myself and some other folks on the steam discussion forums for this game are having similar stuttering and unstable frame them graphs. Can you confirm on your end?

My first impressions of #DeathStranding2PC are very positive, the game looks stunning and runs well overall and so far zero shader compilation stutters or traversal hitches. The graphics settings are more streamlined than typical Nixxes ports but they scale well performance-wise.

@vhsliberal Ninguém te perguntou oh otário

mais um que nem contas sabe fazer.

@pedrouid We don't want them, thanks

If the Portuguese government had more agility... it could've attracted many Dubai expats with a new tax program or visa It’s been almost 3 weeks since the war started... The opportunity is too big to miss and can bring significant capital to Portugal 🇵🇹

@Lx1985_boy @ojeconomico Baterias

@_Fernandes_Yo @ojeconomico e de noite? e quando não há vento? esperamos pelo nascer do sol para ter energia?

Quanto custaria uma central nuclear em Portugal? E qual a potência ideal para o país? O JE entrevistou o professor Bruno Soares Gonçalves, especialista em energia nuclear que fez as contas tendo por base projeto nucleares mais recentes. Saiba mais em jornaleconomico.sapo.pt/noticias/nucle…

@Sebasti66855537 @beyond_fps Turns out this isn't just RR. Enabling RR also automatically increases lighting quality from "Cinematic" to "Max" with no option to decrease it.

@beyond_fps A 2nd-gen Transformer DLSS RR would've probably more than halved the framerate here lmao.

The performance difference enabling Ray Reconstruction is brutal on Crimson Desert, goes from ~80s fps with 4K DLAA to ~55 with DLSS on Quality They must be enabling higher quality RT settings when RR is on, there is no way otherwise

@raywccf @beyond_fps So that explains that DF video. RR alone recovering that much detail was really dubious. (I don't blame Alex on this one, the devs should have been more transparent)

@beyond_fps 1st screenshot is from 4k DLSS 4 Quality with Ray Reconstruction. As you can see, Lighting Quality is set to Max with no further options. 2nd is from 4k DLSS 4 Quality w/out Ray Reconstruction. Lighting Quality is set to Cinematic with one more setting available.

@Joaogalamba @jme20170309 Idem para baterias. Muitas das baterias de carros elétricos ainda funcionam bem, ou aquelas que já não funcionam a uma potência desejável foram reaproveitadas para outros usos como home storage, por exemplo.

@jme20170309 Reciclagem atual é baixa porque ainda não há muitos painéis solares em fim de vida, e ainda não ha muito para reciclar, e porque ser ou não reciclado depende de haver regras que obriguem a reciclagem (as europeias obrigam). Portanto, isto mostra que o texto anterior é falso

Nunca vi o careca da QUERCUS nem o @Joaogalamba falarem sobre isto.

@GhostOfCarroll @Pirat_Nation Certainly more alive than Radeon

@Pirat_Nation Does this mean ARC isn't dead?

Intel has launched precompiled shaders for Arc GPUs. This reduces game load times by up to 3× on average. In some titles, like God of War Ragnarök, the first load can be up to 21× faster. It is now active for 13 Steam games, including Cyberpunk 2077 and Black Myth: Wukong.

@Ice_Majora @OBE1plays

@_Fernandes_Yo @OBE1plays You might be stupid

I BETTER NOT see anyone on this app who complained about dlss 5 today turn it on when you’re playing your games! When you stream, your characters better all look like this 🤣

@ethanh762287 @3DCenter_org Gemini, looking at the fucking watermark they didn't even bother to remove

@3DCenter_org The image itself is ai? They've got lips text block twice. Probably just asked chatgpt to compare.

Analysis: A closer look at DLSS 5 in RE9: Visual differences in base textures vs. DLSS output on Grace's model by Tamerito @ Reddit old.reddit.com/r/hardware/com…

@3DCenter_org This is an AI generated image btw 🤡🤡🤡🤡

@JBU_Show @anton41399829 @FollowTheScien4 @wintonARK Because, as we know, nuclear power has been getting cheaper 😂

@anton41399829 @FollowTheScien4 @wintonARK At the moment.

a barrel of oil can provide as much electricity as a 400W solar panel does annually. a barrel of oil runs $92 and comes with a few minor logistical complications. this year the solar panel should run less than $90; you can order online, ships in a week.

@BrexitDuck @wintonARK I couldn't care less about where it comes from or which energy source it uses

@wintonARK From China built in factories powered by coal fired power stations. Idiot.

@Ice_Majora @OBE1plays Look at a face outside of a cutscene and it will look like trash DLSS 5 makes faces look like a cutscene even during gameplay

@OBE1plays Death stranding 2 btw

@VideoCardz some gamers will genuinely look at you and say that the left picture looks better

Jensen Huang says gamers are ‘completely wrong’ about DLSS 5 backlash videocardz.com/newz/jensen-hu…

@saintjohn69420 @Noahpinion They know what they're doing. If enough of them repeat a lie, some people will believe it by the sheer number of people saying it

@_Fernandes_Yo @Noahpinion Indians in your replies haven't read the paper lol. Neither the op here, but he's just lying for the love of the game.

This paper says India's GDP is overstated by 22%. It also says that Europe's GDP and China's GDP are both overstated by 35%.

@MariaAugustaVi7 Regime antigo de pensões dos banqueiros. É injusto mas é o direito dele.

Como é que isto é possível??? Para um cidadão comum, a idade da reforma é de 66 anos e 9 meses, as pre-reformas estão paralisadas, e este fulano reforma-se aos 59 anos… Que país é este??? 🤬