Search This Blog

Game Consoles Xbox, Microsoft, Play Station, Sony, Killed PC Sound Cards, Sound EAX API, Intel 97, then Intel Azalea, CPU after Pentium 4, Mobile SOC DAC ACD integration tightly, iPod

Sound Blaster made IDE & later PCI sound cards for PC or personal computers in the 1990's

Music Streaming, Computer Gaming, Killer Apps required Motherboards to integrate Sound Processing

When Intel & Microsoft got together to put sound in the PC, this eventually doomed sound cards.

By the time the Play Station launched, first Xbox launched, even PC CPU & GPU CUDA offered high end 7.1 surround sound, elite headphone performance, even over USB-A connections, bidirectional gaming headsets with stereo speakers, synthetic 3D soundscape, and a boom mic to talk smack in games like CS or Counter Strike on Steam, CS2 lately. 

PCIe super fast storage, but Graphics Cards the killer user app for PCI slots in desktops ITX, mATX or full on full size ATX big dog units. 

I am typing this blog posting on an M2 MacBook Pro that has a 1/8 inch multi-function headphone jack.

If we look upon the motherboard and M2 chip of this machine, it features high fidelity audio functions, even over a USB-C connection with high bit rate able to move UHD video, so audio at 198 kilohertz sampling super easy as a simultaneous stream along with ultra high definition video. 

Even my VR headset, a Meta Quest 3S features a Qualcomm SOC with incredible audio performance. 

Everyone's smartphone SOC, system on a chip features high fidelity audio processing, including ADC and DAC for microphone and speakers, noting most phones like the iPhone or Samsung Galaxy S whatever, have a pair of speakers that can pump out stereo sound, from the top earpiece speaker and bottom charging port location speaker, tiny rare earth magnet speakers featuring foam balls to improve sound performance, making the output sound more rich with more base notes and less tin like which would happen without the foam balls improving the sound pressure wave form spectrum. 

When Apple was working with IBM on the first iPod, Steve Jobs contacted Creative in Singapore, to work with them to integrate their sound technology into the iPod line, but Creative declines having their own MP3 player partnerships competing with the iPod, not seeing the sales potential of Apples music distribution AAC files on iTunes online, which came to dominate music sales and the CD players faded into tech history, discs physical kept alive with DVD and lastly BlueRay disks, but no new DISC formats after that, especially not for music alone. 

Something about writing all of this reminded me of the Microsoft Zune audio player & software. 

I don't even remember the name, but I had an MP3 player in college that used CF Compact Flash Memory Card, it was a Canadian Brand, made in Singapore, around 2001, Napster online download many gigabytes of MP3 music, over USB loaded them onto this portable CF card MP3 player. 

2026 Expensive PC Parts Forecast of Higher Prices Upcoming

DDR5 6000MHz RAM for PC, $400

5090 GPU $4000

i9 or Best Ryzen $1000

High end 1000w PSU 80+ Titanium $400

Liquid Cooling for CPU & GPU + radiators, hoses, additives for fluid, kit, $1000

Your looking at a high-end gaming PC for $6000 that blows any gaming laptop out of the water in terms for FPS on AAA titles

Many high-end gaming laptops are selling for $2500-$4500 now

Even popular gaming consoles XBox X and latest PS5 prices are up to $800

AI demand for GPU, RAM, CPU, logic, all the part sharing the same supply chain as PC parts

Leveraging more than $700 billion in ai investments 2026, the ai industry making PC parts expensive

Graphics Cards are Super Computers

 Even very cheap GPU cards are pumping data at 2 teraflops, trillions of floating point calculations per second, often on about 20 watts of power, while even the 70w RTX 3050 6GB OC does just above 6 teraflops.

A 4060 does about 12 gigaflops of floating point, on a larger power budget of 150 watts, requires a PSU with a GPU power cord able to give the GPU enough energy, noting that a GPU contains VRMs to regulate RAM & GPU power, a mainboard logic, data controller for IO, and other components of a PC motherboard, and in some sense a gaming GPU card a kind of specialized super computer. 


The same can be said of the latest updated Sony Play Station 5 game console or the Xbox X game console, which are gaming computers with highly optimized architectures. Even the Intel Arc 580 with 12GB VRAM performs well at the $450 price point to keep up with vastly more expensive NVIDIA chip GPUs

AI boom eating up global RAM production so much that RAM today sells for 4X what it did 1 year ago. I had to pay almost $240 for 4X8GB of DDR4 3200MHz w/ Aluminum Heat Sinks, while the 3050 6GB OC GPU card cost me $180, the MSI PRO B550-VC mainboard was $80, the AMD Ryzen 5-5500 was $86, the Zalman CUBIX-G case for $40, the Rosewill 650W 80+Gold PSU was $47, 22inch Gaming Monitor with speakers was $70, gaming keyboard was $10, all said with a DP braided cable, about $810


Anthropic, OpenAI, SpaceX, wait a Rocket Company big on Artificial Intelligence

Lets not forget Apple Siri or Google/ Alphabet Gemini or Amazon Alexa, notice how even Microsoft Copilot/ Cortana/ Clippy in Microsoft Office, or has narrow band ai been present in IDE SDE when writing code the compiler used specialized ai to check code, and many writing applications have spelling and grammar checking built in, also narrow band type of artificial intelligence, and the same can be said of the many generations of bot crawlers the browse and scrape the web to build a search index ranking key words list and similar, essential for companies to connect with customers online, search engines. 

So why would SpaceX go in big on ai development ? Lets not forget the FSD packages sold on Tesla model whatever, which a 10 year delayed promise level 5 autonomy never delivered, causing many to sell their Tesla Model 3, S, X, Y, or Cyber-truck, oh noting that LIB EV's like those Tesla mentioned, the same thing that gives them really fast acceleration, also shreds tire tread up to 10X faster than ICE or conventional gasoline engine powered automobiles. 

Tesla ^ SpaceX are the same people, so ai development essential for Tesla FSD can aid rocket guidance system development at SpaceX

Notice fuzzy logic controller of a typical automatic transmission of an automobile, SUV or Truck, this high speed analogy circuit a kind of very low brow very narrow band ai, as it does everything automatically without using a big energy hogging digital multipurpose computer module, like the ECU or VCU. Notice also that nearly every kind of passenger car sold contains up to 100 chips, integrated circuits, similar to the SOC or brains of a smartphone. 

IO, EUROPA, MARS, MOON, SUN, SpaceX, Elon Musk, Chemical Rocket Mass Paradox, Urgent Need for Space Based Fusion FRC Low Mass Propulsion

United Launch Alliance, Reusable Rockets, Solid Rocket Boosters, 3S Rockets, Space


Amazon New Shepard Explosion on vs Kepler Satellite Internet Amazon.com Prime


Star Link space based high speed internet satellites + ground receiver kits from SpaceX


Global AI race chip GPU RAM demand ai training paradox cost, more electricity & time

New algorithms taking longer & costing more to develop, scraping internet data 

Big data explosion generated content videos impossible scenes depicted

Text prompts making images or videos automatically, data centers cost billions

PC parts graphics cards, M2 MVNE SSD, RAM, CPU price debacle 2026 So Expensive

NVIDIA and Microsoft for new kind of CPU, via GPU skills for WINTEL history technology

E-commerce, Investment, Fintech, Crypto, Blockchain, Insane in the Brain Membrane 

Catch that riff, different, slippery slope, hope for it, spit it down, sound of which, switch


Twitch streaming videos YouTube Amazon Prime Video, Netflix, Pick up sticks, wicks 

Legal tender fiat currency global national debt debacle increasingly unstable able to pop

Bubble gum, dumb as a rock, sock missing, hissing sound indicates a leak, take a leak

Pop! OS by System 76, Firefox Web Browser for Low-end Gaming PC Build, Ryzen 5500, B550 MSI, GTX 3050 6GB, 32GB DDR4 3200Mhz, 650W 80+Gold PSU, Vulcan 256GB M2, Still cost me $800.48

Though I got a full gaming keyboard, dope monitor, nice case with 3 fans included, a really novel CPU cooler, really high end PSU, dope mainboard, & what appears to be good ram with aluminum heat sinks, through I dreamt of smashing some copper water pipe to make a custom heatsink for the VRM's on the mainboard and GPU, to keep them a tad cooler to make them last longer. Relatively easy, it gives me a path to some DIY fun, as does some bright yellow spray paint in a can that I will use to custom paint part of my rig to personalize it, including graphics for Pop! OS by System 76, Firefox Browser, Wifi 6 MSI B550, GTX3050 6GB OC, 32GB 3200MHz DDR4, 650W 80+Gold PSU, Up-cycled M2 drive 256GB


AMD Ryzen 5-5500, B550 MSI PRO VC, ASUS dual fan GTX 3050 6GB, 32GB DDR4 3200Mhz Puskill Gaming RAM, Rosewill 650W 80+Gold PSU, Vulcan 256GB M2, Still cost me $800.48


I harvested M2 Vulcan 256 from an old mining rig, while 64GB Kingston USB key used to flash the ISO was found on a public bus with nothing on it. Waiting on Amazon for all the parts, but decided on System 76's Linux as the best net solution for CS2 on PC, but also allowing Web with Firefox secure and lightweight browser with low ram footprint. The 70w low power GPU can run on PCI-e 4.0 port DC energy budget of 150 watts, while this build allows for a better graphics card, as the PSU has a dedicated 450 watt cable for GPUS, but that will be retained for a future upgrade. Desktops being full modular, almost like Lego's to put together, can be updated, mainboard, ram & cpu, gpu, PSU, case, cooling method, its all parts that go together in millions of possible combinations. Why did I go with GTX 3050 6GB OC Asus Dual Fan, because it was $180, while the 4X8gb DDR4 32GB 3200MHz cost $232, more than $50 more than the GPU, noting RAM prices are high because of demand for AI training, but odd that DDR4 288pin desktop sticks would be expensive, who uses those on AI training rigs anyways?