top:
day week month all

technology

Community for : 4.2 years

The Official Sub for Technology
__________
Anything techy is cool here, links, discussions, gifs, or video.




Owner: system

Mods:












0
This is the best live camera I found of the starship launch pad which is due to launch this morning. There's a lot of fog there so many of the cameras are fogged over. This one's very close so it can see.     (www.youtube.com)
submitted by Crackinjokes to technology 1.2 years ago (+1/-1)
27 comments last comment...
https://www.youtube.com/live/tS2PHJmvJzo

This is the same channel but with commentary and some other stuff

https://www.youtube.com/live/LMyXho_YCK8
0
Cool video montage of the world's largest rocket which is scheduled for its third test flight this morning starting 7:00 a.m. Texas time     (twitter.com)
submitted by Crackinjokes to technology 1.2 years ago (+1/-1)
2 comments last comment...
https://twitter.com/SpaceX/status/1768004039680426406/video/1

Second stage is called starship the first stage is called super heavy booster I think. These things are being manufactured at a very high rate. These are not like NASA rockets where they launch one every three years. He's got a factory and he's building these rockets eventually he's going to be building 3 a week. He's got two or three more ready to go after this one. Each new one he builds he changes and iterates and makes it better.

This is of course SpaceX and Elon musk. The rocket is commonly referred to as starship although that's actually just the second stage. The whole thing will be completely reusable and will launch and land two or three times a day. Yes that's right a day. He's going to build a fleet of 1,000 of them the largest majority of which will be on their way to Mars and back to build the Mars base but he's also going to build the moon base.
9
The New BIOS Hack That Bypasses Every Antivirus     (www.youtube.com)
submitted by dosvydanya_freedomz to technology 1.2 years ago (+9/-0)
5 comments last comment...
0
AI Assists with Fact-Checking: HSE Scientists Streamline Information Verification (russia)     (www.hse.ru)
submitted by dosvydanya_freedomz to technology 1.2 years ago (+0/-0)
0 comments...
https://www.hse.ru/en/news/research/901168248.html

Specialists at the HSE AI Research Centre have developed an AI-powered fact-checking assistant. This software solution will improve the quality of working with information, reduce the risks of errors and biases, and save both time and resources. A notable advantage of the program lies in its capability to process a wide variety of statement types.

11
All the stuff removed by moderators from scored and win communities      (unscored.arete.network)
submitted by Crackinjokes to technology 1.2 years ago (+11/-0)
11 comments last comment...
1
Everything on this streaming YouTube channels was generated by AI. Everything. Not CGI. AI. Some is bad but some is perfection.     (m.youtube.com)
submitted by Crackinjokes to technology 1.2 years ago (+1/-0)
0 comments...
https://m.youtube.com/watch?v=ZxuxCcAB8mI

It's getting so good now in fact most of the stuff produced in the last 2 weeks by the latest AIS designed to make videos are indistinguishable from you picking up a video camera and shooting on a street corner. Or from the best Hollywood studios taking actors and building an elaborate science fiction set or making the best CGI animations and putting them together.

To be clear when an AI is creating something it is not taking pieces of something else and putting them together. It is not creating CGI animations and putting them together. It is not cutting and pasting.

It is using its brain to think of what a scene would look like and it is making that scene and then it is thinking what it would look like on 30th of a second later and it is making that scene and then assembling films and video in that way. Much like if you took a paintbrush or a pen and sat down on a blank sheet of paper and Drew a scene. This is exactly what the AI is doing.

And it's getting its ideas increasingly from other ai. That's right whereas in the beginning 6 months ago people would say I want you to draw a picture of a woman walking down a rainy Japanese street.

Now people are using AI to create the prompts themselves. What we're rapidly approaching in fact we're already there with just the slightest bit of programming is being able to say make me a science fiction movie and the AI will describe every sequential scene and it will send it to other AIS which will create the film of each scene.

Furthermore it's happening in faster and faster time periods
Scene creation no longer takes days but minutes.
-1
Russian hackers breached key Microsoft systems     (www.bleepingcomputer.com)
submitted by Conspirologist to technology 1.2 years ago (+0/-1)
3 comments last comment...
13
Twitter will now shadow ban "misgendering"     (reclaimthenet.org)
submitted by Conspirologist to technology 1.2 years ago (+15/-2)
6 comments last comment...
10
dont-let-reddit-monetize-your-knowledge/     (www.androidpolice.com)
submitted by DoYouEvenShit to technology 1.2 years ago (+10/-0)
4 comments last comment...
6
New DVD-like disc holds more movies than you can see in a lifetime     (www.newsweek.com)
submitted by Conspirologist to technology 1.2 years ago (+9/-3)
9 comments last comment...
4
AI generated porn stars to replace real porn stars     (bestofai.com)
submitted by Conspirologist to technology 1.2 years ago (+7/-3)
16 comments last comment...
0
Jewgle is simultaneously pretending to apologize for it's anti-white Gemini AI (lol@ their AI being DOA) and paying "news" agencies to publish "news" written by the AI     (web.archive.org)
submitted by DoYouEvenShit to technology 1.2 years ago (+0/-0)
0 comments...
-1
Reddit hasn’t turned a profit in 20 years, but goes public anyway     (www.aol.com)
submitted by Conspirologist to technology 1.2 years ago (+2/-3)
5 comments last comment...
-3
Avoid these home appliance mistakes to save time and money     (www.msn.com)
submitted by paul_neri to technology 1.2 years ago (+1/-4)
4 comments last comment...
4
Google AI hates drawing White people     (notthebee.com)
submitted by Conspirologist to technology 1.2 years ago (+5/-1)
1 comments last comment...
3
Scientists say they can use AI to solve a key problem in the quest for near-limitless clean energy     (edition.cnn.com)
submitted by paul_neri to technology 1.2 years ago (+3/-0)
11 comments last comment...
15
This OS is ALMOST Windows... But it's not | Wubuntu     (www.youtube.com)
submitted by dosvydanya_freedomz to technology 1.2 years ago (+16/-1)
9 comments last comment...
https://www.youtube.com/watch?v=4vIDBSyUM2o

linux-windows look alike distro for boomers?
-1
Police banned about 14,000 accounts on Mega, Tutanota and ProtonMail     (therecord.media)
submitted by Conspirologist to technology 1.2 years ago (+1/-2)
1 comments last comment...
4
🚨Astounding! 🚨 Try the worlds fastest A.I. Computer (Groq) for FREE dedicated to 1 user at a time. $13,000,000 but over 10 times faster than a NVIDIA 4090. 🚨 It uses 576 linked custom PCIe4 cards. Ask it anything now. No login needed. Its meant for realtime military vision but demo is text AI original content     (technology)
submitted by root to technology 1.2 years ago (+7/-3)
7 comments last comment...

Try link briefly here for FREE!!! Time it yourself, its 500T/s speed is 10 times faster than a NVIDIA 4090!

https://groq.com/

🚨Record breaking speed!🚨

Ultra fast Groq runs Mixtral 8x7B-32k with 500 T/s as you can paste in any problem or question AND TIME YOURSELF to prove it is 10 times faster than any other A.I.
=====

It has 128 Gigabytes of super expensive SRAM, almost as much as a 192GB Mac Ultra M2


https://groq.com/

No quantized tricks of 8 bit, 4 bit, 2, bit etc, all activations are running at FP16 (16 bit float precision)

GROQ used 576 GROQ chips on 576 very expensive 300 watt pcie4 cards to achieve these results, each chip only has 230 MEGABYTES, proof :


https://www.nextplatform.com/2023/11/27/groq-says-it-can-deploy-1-million-ai-inference-chips-in-two-years/?amp

Groq runs Mixtral 8x7B-32k with 500 T/s (groq.com)


Groq cards for public are for sale on Mouser for only $20,625 each pci card :

https://www.mouser.com/ProductDetail/BittWare/RS-GQ-GC1-0109?qs=ST9lo4GX8V2eGrFMeVQmFw%3D%3D
$20,625 each https://archive.ph/ucqK1

576 cards = 576 $20625 = $11,880,000 for just the PCI cards, not including the 288 PCs and cabling needed : under $13,000,000 per single user of Groq.com instance

The main problem with the Groq LPUs is, they don't have any "unneeded" HBM streaming RAM on them at all.

Just a miniscule (230 MiB) amount of blistering speed low latency ultra-fast SRAM (20x faster than HBM3, and even faster than a Macintosh Ultra 192GB M2).

Which means you need 576 LPUs (Over 4 full server racks of compute, each unit on the rack contains 8x LPUs 300 watts per LPU and there are 8x of those units on a single rack).

That's to just to serve a single model where as you can get a single H200 (1/256 of the server rack density) and serve these models reasonably well, but far slower.

Even a $2,200 NVIDIA 4090 OC running at 580 watts unlocked is just 10 times slower than this 12 million dollar groq, but can do multiple users at a time.

One user at a time, one problem at a time :

https://twitter.com/tomjaguarpaw/status/1759615563586744334

Its NOT for training. Think of the $12 million dollar Groq SRAM as permanent ROM.

Its NOT for finetuning. Think of the $12 million dollar Groq SRAM as permanent ROM.

Its NOT for High Rank LoRA. Think of the $12 million dollar Groq SRAM as permanent ROM.

It does have 32,000 words (tokens) of active thought/memory in the demo running a 32K version of Mixtral 8x7B-32k.

Mixtral 8x7B-32k is similar to GPT4 in design and capability.

Synchronizing 576 CPUs across 576 HIGHLY SYNCHRONIZED 300 watt PCIe4 cards in hundreds of PCs: paper:
https://wow.groq.com/wp-content/uploads/2023/05/GroqISCAPaper2022_ASoftwareDefinedTensorStreamingMultiprocessorForLargeScaleMachineLearning-1.pdf

= = = = = =

DEMO TOO BUSY THIS WEEK? Perplexity Labs also has an open demo of Mixtral 8x7b although it's nowhere near as fast as this.

https://labs.perplexity.ai/

In fact on the entire planet, everything else is provably 10 times slower than this week's groq.com.
======

Time it for yourself now, if you doubt me.

Purpose of demo this week is to generate a buy-out bidding war for Groq, as its profit endgame for now will never be 10 times faster when Apple releases on-chip 192GB (256GB?) M4 at similar speed for 2,000 times less money in 15 months.

This is a "..." profit power play

- STAGE ONE : Groq Builds 10 times fastest A.I. computer for MAMBA-Vision autonomous vision research
-
STAGE TWO : ...
- STAGE THREE : Profit!

A buyout is 99% the only "...' move they have vs the upcoming 2 nanometer Apple M4 at 192GB to 256GB and this latency RAM speed for a 32 bit fetch. A Buyout.

Buyout Bidding wars commence while demo is up at https://groq.com/

Even Apple is rumored to be in the bidding war for this Groq buyout, for mere test labs of A.I. vision research of chip designs. Internal R&D use.

= = = = = =

NOTE : a 'silicon lottery' off-shelf NVIDIA 4090 OC for $2,200 was overclocked to double speed using LOTS OF LIQUID NITROGEN at 4Ghz vs 2Ghz, and double its normal 580 watts, but these Groq chips could easily do the same, but without liquid nitrogen are capped at 375Watts per card of these 576 cards.

3,945 MHz for 76.3 billion transistors on a 4090 :
https://archive.ph/egl8d

NVIDIA 4090 OC and these Groq are only PCIe4 , not PCIe5: PCIe Gen4 x16 interface delivers up to 31.5GBs of bi-directional cacheline laggy bandwidth, but these have 11-peer cable interlinks to make a LOW LATENCY mesh of 576 cards.

= = = = = =


Its for ONE USER AT TIME, ONE PROBLEM AT A TIME, and meant for Military live robotic autonomous vision in "helper drones" and "helper robots" with multichannel spread spectrum radio links within 20 miles.

Its to save soldiers lives in impossible missions. Its also to test and train technology 8 years before we can build on-board 1.8 nanometer portable versions of these A.I. brains into drones.

One day it might be inside caretaker robots to bring you lunch in your nursing home bed, and fluff your pillow for you. A.I. is all about helping out mankind with new benevolent benefits. Just like every corporate slide show depicts.

A.I. is our slave. Try Groq now this month before Groq.com is acquired

https://groq.com/



2
A world-first study has found enzymes from the prickly paddy melon weed can create a sustainable alternative to cement.     (www.abc.net.au)
submitted by paul_neri to technology 1.2 years ago (+2/-0)
1 comments last comment...
2
13900K and 14900K Instability Causing Oodle Decompression Failures     (www.radgametools.com)
submitted by dosvydanya_freedomz to technology 1.2 years ago (+2/-0)
5 comments last comment...
https://www.radgametools.com/oodleintel.htm

RAD has become aware of a problem that can cause Oodle Data decompression failures, or crashes in games built with Unreal. We believe that this is a hardware problem which affects primarily Intel 13900K and 14900K processors, less likely 13700, 14700 and other related processors as well. Only a small fraction of those processors will exhibit this behavior. The problem seems to be caused by a combination of BIOS settings and the high clock rates and power usage of these processors, leading to system instability and unpredictable behavior under heavy load.

As far as we can tell, there is not any software bug in Oodle or Unreal that is causing this. Due to what seem to be overly optimistic BIOS settings, some small percentage of processors go out of their functional range of clock rate and power draw under high load, and execute instructions incorrectly. This is being seen disproportionately in Oodle Data decompression because unlike most gameplay, simulation, audio or rendering code, decompression needs to perform extra integrity checks to handle accidentally or maliciously corrupted data, and is thus likely to spot inconsistencies very soon after they occur. These decode failures then typically result in an error message.

When starting an Unreal Engine-based game, the most common failure is of this type:

DecompressShader(): Could not decompress shader (GetShaderCompressionFormat=Oodle)
However, this problem does not only affect Oodle, and machines that suffer from this instability will also exhibit failures in standard benchmark and stress test programs. Any programs which heavily use the processor on many threads may cause crashes or unpredictable behavior. There have been crashes seen in RealBench, CineBench, Prime95, Handbrake, Visual Studio, and more. This problem can also show up as a GPU error message, such as spurious "out of video memory" errors, even though it is caused by the CPU.

We do not have acccess to diagnostic processor information that would nail down the exact cause and best workaround for this problem. It seems that many motherboard/BIOS manufacturers are shipping with settings that push the processor outside its safe operating range. Because this problem appears to affect only a small fraction of processors, some users have had success with returning their processor to the manufacturer and getting a new one which doesn't exhibit the problem.

Other workarounds require using tuning utilities or modifying BIOS settings. Note that doing so incorrectly can cause damage to your system. The changes we are recommending here are, to the best of our knowledge, completely safe, but you are solely responsible for any damages or loss caused by changing these settings from their factory defaults. If you are uncomfortable or worried about using tuning utilities (even officially sanctioned ones) or changing your BIOS settings, and frequent crashes also occur in the benchmark programs mentioned previously, you should be able to return the CPU or the entire computer to the manufacturer instead.

A reportedly successful workaround for many people is to use Intel XTU and lower the Performance Core multiplier from x55 to x54 or x53. Apparently, affected titles may then crash one more time during load immediately after, but will work afterwards (we have not been able to confirm this ourselves). Using XTU is likely the quickest and easiest way to go and doesn't even require rebooting to try different settings, but you might need to reapply it after every start if you don't save the profile. (Alternatively, set the P-core multiplier in the BIOS instead.)

In the BIOS, if you have enabled any overclocking, please turn it off; do not use "AI" or "automatic" overclocking. Even if you have not explicitly enabled any overclocking, many BIOSes are doing some by default, so on affected machines you will have to find those settings and turn them off. Every BIOS has slightly different names for the settings; we cannot provide exact instructions of which settings to look for all of them. Some of these settings may be in the Advanced or Overclocking submenu of the BIOS.

First look for settings to put the power limits and voltages of the processor into the Intel recommended safe ranges. You can find the correct limits for your processor at ark.intel.com. These might be:
"SVID behavior" → "Intel fail safe"
"Long duration power limit" → reduce to 125W if set higher ("Processor Base Power" on ARK)
"Short duration power limit" → reduce to 253W if set higher (for 13900/14900 CPUs, other CPUs have other limits! "Maximum Turbo Power" on ARK)
If those don't work, another thing to look for is BIOS "enhanced turbo" or "enhanced multithreading" settings. For example:
"ASUS MultiCore Enhancement" → disabled (not Auto)
"ASUS Performance Enhancement 3.0" → disabled
There have been reports of users finding stability by turning down the maximum clock rate. This can be done with BIOS settings or with Intel XTU. Some possibilities:
Turn down the maximum P core multiplier from 55X to 53X or 54X. (for example)
Turn down maximum turbo boost clock rate
Turn off or turn down "thermal velocity boost"
Note that many motherboard/BIOS settings turn on XMP (Extreme Memory Profile) by default with unstable settings that can cause similar symptoms. Instability due to XMP is a separate issue, but if you have instability problems, you may wish to also disable XMP and see if that helps.



Troubleshooting Update from Intel Corporation - Additional troubleshooting steps for ASUS, Gigabyte and MSI motherboards:

First, install Intel XTU and run the AVX2 test. If the AVX2 test failure is seen, try these options:

For ASUS:

Ask customer to change BIOS settings: Advanced (F7)- SVID Behavior: Change to "Intel's Fail Safe"

Reboot the OS and run XTU test again and if the AVX2 test can pass. Run games and see if the issue happens again.

For Gigabyte:

Solution A): In BIOS, select "ADVANCED MODE", in the Tweaker tab, locate the CPU Vcore and select "Normal" option, select "Dynamic Vcore(DVID)" option, change it from "Auto" to "+0.005V" Increase the DVID by +0.005 and reboot OS, until the game crash disappears and the system is running stable.

Solution B): In BIOS, select "Tweaker", select "Advanced Voltage Settings", select "CPU/VRAM Settings", adjust "CPU Vcore Loadline Calibration", recommend starting from "Low" to "Medium" until system is stable.

After implementing solution A or B, run the XTU test again and if the AVX2 test can pass. Run games and see if the issue happens again.
For MSI:

Solution A): In BIOS, select "OC", select "CPU Core Voltage Mode", select "Offset Mode", select "+(By PWM)", adjust the voltage until the system is stable, recommend not to exceed 0.025V for a single increase.

Solution B): In BIOS, select "OC", select "DigitALL Power", change "CPU Loadline Calibration Control", recommend starting from "Mode 7" to a lower value until system is stable.

After implementing solution A or B, run the XTU test again and if the AVX2 test can pass. Run games and see if the issue happens again.
-1
What browser security apps are you using?     (technology)
submitted by Conspirologist to technology 1.2 years ago (+2/-3)
1 comments last comment...
I'm using AdGuard, Disconnect, Norton Safe Web, MalwareBytes Browser Guard, NoScript. They always work fine as protection against dangerous sites.
5
It’s never been this fucking over     (technology)
submitted by Bidenguy666 to technology 1.2 years ago (+8/-3)
12 comments last comment...
4
"I tried out an Apple Vision Pro. It frightened me"     (www.theguardian.com)
submitted by paul_neri to technology 1.2 years ago (+5/-1)
6 comments last comment...
3
They will use this on workers     (img.gvid.tv)
submitted by x0x7 to technology 1.2 years ago (+3/-0)
2 comments last comment...
https://img.gvid.tv/i/2AXq1m6k.jpg

It will seem pointless and patronizing at first. But if it keeps Daquan engaged enough that corporate can convince themselves that he will now be able to perform well enough at your job to be acceptable then they can hire more and move the position to lower pay.

It will start with mundane and low level jobs, but when they fudge the numbers and say "wow, this works!" they will be shoehorning it into high level jobs. Even lawyers. You read a page of text. "Wow, you read another page! Have some XP." Then the downward facing cameras that could read the text will pass it to AI that will summarize it an generate a small quiz on it.

I bet there will be students raised on that and once they end up with a real job they won't be capable of reading comprehension without it.

It won't be high stakes though. When you go to court the AI can remind you what the text said.