dubito, ergo sum, vel, quod idem est, cogito, ergo sum
Joined April 2022
- Tweets8.2K
- Following4.9K
- Followers726
- Likes43.4K
Pinned Tweet
Looking at my account and my TL?? Don't forget to follow me....
Warning: I can be weird sometimes....
LeDarthPlanet retweeted
NVS-03 launched successfully last night on GSLV-F18.
The payload that actually matters for the long term is the third indigenous Rubidium Atomic Frequency Standard (iRAFS / IRAFS) developed at SAC Ahmedabad.
This is not just a normal clock.. it's the core of the navigation..🧵
Notice the wordings of these two joo-dicksucker goyims and connect the dots....
I agree with Sam Altman on this.
Running local models sounds appealing in theory, but the reality is underwhelming. Most are still too slow and lack the intelligence to be genuinely useful for serious work.
You could easily spend $1,000+ building a machine with 128GB of RAM, only to end up running a model that delivers mediocre results at 8 tokens per second.
Meanwhile, a $20/month subscription to a cloud-based AI service can get significantly more actual work done.
There's also the issue of resource utilisation. Your expensive RAM sits idle most of the time, whereas data centres pool resources and utilise infrastructure far more efficiently.
Owning the hardware doesn't necessarily mean getting more value out of AI.
Northeast be getting HSR connectivity before South and East but sure....
BJP/"Mainlanders" doesn't care about us Northeast people....
#WATCH | Varanasi, UP: Union Railway Minister Ashwini Vaishnaw says, "Prime Minister Narendra Modi has approved the 'Northern Corridor' of the bullet train network. This will begin in Delhi, connecting Agra, Kanpur, Lucknow, and Ayodhya, and will extend further through Varanasi and Patna, reaching all the way to Guwahati. Once this Northern Corridor is completed in the coming years, you will be able to travel from Delhi to Agra in just 59 minutes; Delhi to Kanpur in one hour and 45 minutes; Delhi to Lucknow will take two hours and 12 minutes; Delhi to Varanasi will take 3 hours and 33 minutes and to Prayagraj, 33 minutes... Many people—especially those in the opposition—will repeatedly spread the misconception that this bullet train is only for the rich... The fare will be very economical and much less than flight tickets..."
LeDarthPlanet retweeted
One thing i dont get from the Indian Defence X community, almost all of whom r pronouncing TEDBF dead bcoz French Rafale started flight test on Vikrant....not the rafaleM v had ordered, which themselves wil take another 4-5yrs for their delivery to complete.Thats 2031.
Even while F/A-18 & Rafale r being tested for selection to the time RafaleM deal was signed in 2025, TEDBF design kept evolving ...and they showed 1:4 model in last Aero India, alognwith timelines .Just last oct, Navy chief himself had said "TEDBF is an important project for the Navy"..But somehow, IDX "celebrates" orbituary of TEDBF every month. If ADA-DRDO-HAL-Navy kept working on TEDBF and kept refining the design to current one,despite of rafalM deal was being negotiated and signed, where did this news of TEDBF getting killed come from ?
Is it their inherent opinion or is there an official confirmation or Navy chief conirmed ?
New toolkit for AI bros in India:
> Create brand new, insta 2k followers
> Set location to the most controversial part of the country, while living in slums
> Put out takes worse than RaGa and Dickpe
> Suck jewish cock and preaching "ownership is a crime"
Jaha dekho jor se pelo
I Agree with Sam Altman.
First of all, local models are useless, they're dumb and slow
You'll spend ~$1000 on 128 GB RAM to run a local model
And you won't get anything done with it, dumb models at 8tps are useless
A $20 subscription will get more things done
And for 95% of the time, RAM is not being used so it's doing less work there compared to in a data centre where utility is more
Readers added context they thought people might want to know
Local models on suitable 128GB hardware often run at 30+ tokens/second and offer practical uses; 128GB DDR5 RAM kits currently start above $2200, not ~$1000.
datahardware.ai/tools/benchmar…
theramageddon.com/128gb-ddr5-ram…
Many will notice one of the founders of one pseudo-Indian (American) company has now changed his pfp to one with Modiji to showcase some sort of patriotism??
Respect for you and your work, but don't get benefitted from govt with "Indian" tag and make home in and serve the USA.
Pray tell me why the cost of 128GB RAM is $1000 now, and is RAM only for AI instead of, you know, using my mobile and computer in regular way, playing games, doing projects and stuff, you jew-dicksucker??
I Agree with Sam Altman.
First of all, local models are useless, they're dumb and slow
You'll spend ~$1000 on 128 GB RAM to run a local model
And you won't get anything done with it, dumb models at 8tps are useless
A $20 subscription will get more things done
And for 95% of the time, RAM is not being used so it's doing less work there compared to in a data centre where utility is more
Readers added context they thought people might want to know
Local models on suitable 128GB hardware often run at 30+ tokens/second and offer practical uses; 128GB DDR5 RAM kits currently start above $2200, not ~$1000.
datahardware.ai/tools/benchmar…
theramageddon.com/128gb-ddr5-ram…
LeDarthPlanet retweeted
India's semicon push should encompass all major types of logic & memory devices needed for consumer retail applications & not just enterprise use.
Profitability may be lower but this market is never going to vanish or rather shouldn't be allowed to be destroyed.
Perhaps greater incentives can be provided, a few years from now India should be among the very few countries with factories still manufacturing logic & memory for retail customers.
FPT Cursor engines ahhhh...🤤🤤🤤
Endless possibilities ahhh....🤤🤤🤤
Now any real tech infusion into Tata Motors (perhaps even India's defence sector) has to be via FPT Industrial which is the Iveco unit that deals with powertrains:
- 6 engine ranges: 30 - >1,000hp
- natural gas engines: 50 - 520hp
- Hythane, hydrogen, hybrid, electric
- transmissions upto 500Nm torque
- front & rear axles: 2.45-32 tons GAW
Well worth the price if Tata can manage & utilize well:fptindustrial.com/en/Find-Yo…
LeDarthPlanet retweeted
Om to the pilots but something need to be addressed
All the people who were crying about crash of ALH as a justification for H125 are very silent now
Hell Olivier Dassault the French MP and billionaire also died in a H125 crash
2 out of 4 Hermes 900 ordered by india has crashed so far and people don't have any issue with it
But the moment any domestic product say Tapas or Tejas crashes they become the most unreliable aircraft in the whole world
LeDarthPlanet retweeted
I hope that @AshwiniVaishnaw and @RailMinIndia are following this accident very closely. This accident is the manifestation of a systemic breakdown in the overall operation of Spanish high speed system with a combination of both mechanical, electronic as well as human errors each building up one after another aligning themselves like the holes of a Swiss cheese for the accident to occur. After this I would seriously review the safety systems being contemplated for our own forthcoming high speed and semi high speed railway networks with particular attention being paid to redundant systems that will ensure that if there is any breakdown or failure detected, then all trains will come to a stop, regardless of what the OCC or the loco pilot decide upon. A hard, inconvenient stop and shut down is far more desirable than playing with the lives of hundreds if not thousands of passengers any day.
Inventor of Graphene BC 😳😳😳
@pk00202500
Nobel Laureate Prof. Sir Konstantin Novoselov joins hands with LOHUM as Chief Scientist of our new incubation & engineering platform for advanced materials.
#AdvancedMaterials #ViksitBharat
