Загрузка видео...

Не удалось загрузить видео

На главную

To build a generic swap in C, pass the size explicitly, create a temporary byte buffer, then use memcpy three times to rotate the bit patterns. VLA char buffer[size] is a GCC thing, for a more portable version malloc the temp and free it. Stanford CS107, lecture 4

tetsuoai's profile picture

tetsuo

237,066 subscribers

45,838 просмотров • 24 дней назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

You cannot add to a void pointer in C. The compiler does not know the size. The everyday fix is to cast it to char* so the math is in bytes, then multiply by the element size yourself. Stanford CS107, lecture 4
1:37

You cannot add to a void pointer in C. The compiler does not know the size. The everyday fix is to cast it to char* so the math is in bytes, then multiply by the element size yourself. Stanford CS107, lecture 4

tetsuo

25,772 просмотров • 25 дней назад

In C, `array[i]` does not add i to the address. It adds i times the size of one element. Assembly has to do that multiply. C hides it. Stanford CS107, lecture 9.
0:44

In C, `array[i]` does not add i to the address. It adds i times the size of one element. Assembly has to do that multiply. C hides it. Stanford CS107, lecture 9.

tetsuo

69,736 просмотров • 25 дней назад

** Sega Genesis 3D engine update 4 ** Never can leave a good rabbit hole undug ! .. updates since last update : Changed frame buffer from a VDP tile friendly buffer to a linear buffer which needs CPU translation , which speedup drawing a LOT not having to worry about tile boundaries albiet slows buffer sending to VRAM. The interior filling of triangles now is running at 1.5c ( 68k cycles ) per pixel ! however sending the buffer is twice as slow per byte as we need the CPU to send the buffer to VRAM to write to the tile format the VDP needs, rather than DMA directly to VRAM . Overall speedup 25% + as the drawing speedup is much greater than the frame buffer slow down. Optimised Line drawing and Triangle filling -near optimal now for the linear frame buffer. Triangle drawing might get a little bit more yet. Quad drawing is next on the list as that will speedup drawing quads. Some gains will come from been smart about partial clearing of RAM buffers and partial sending to VRAM buffers, rather than send the whole buffer. Vertice rotation and projection needs a lot of optimising .. i expect good gains to come from this - this is all in C at present and all calced in real time. Line and Triangle drawing are in assembly but all vertex manipulation is in C. #SGDK #MegaDrive #SegaGenesis
0:24

** Sega Genesis 3D engine update 4 ** Never can leave a good rabbit hole undug ! .. updates since last update : Changed frame buffer from a VDP tile friendly buffer to a linear buffer which needs CPU translation , which speedup drawing a LOT not having to worry about tile boundaries albiet slows buffer sending to VRAM. The interior filling of triangles now is running at 1.5c ( 68k cycles ) per pixel ! however sending the buffer is twice as slow per byte as we need the CPU to send the buffer to VRAM to write to the tile format the VDP needs, rather than DMA directly to VRAM . Overall speedup 25% + as the drawing speedup is much greater than the frame buffer slow down. Optimised Line drawing and Triangle filling -near optimal now for the linear frame buffer. Triangle drawing might get a little bit more yet. Quad drawing is next on the list as that will speedup drawing quads. Some gains will come from been smart about partial clearing of RAM buffers and partial sending to VRAM buffers, rather than send the whole buffer. Vertice rotation and projection needs a lot of optimising .. i expect good gains to come from this - this is all in C at present and all calced in real time. Line and Triangle drawing are in assembly but all vertex manipulation is in C. #SGDK #MegaDrive #SegaGenesis

Shannon Birt

15,280 просмотров • 4 месяцев назад

.Lukky showcased the power of Godot 4.5's stencil buffer with a cool holographic display and shared the project for free. Check it out and learn more about how the stencil buffer works:
0:22

.Lukky showcased the power of Godot 4.5's stencil buffer with a cool holographic display and shared the project for free. Check it out and learn more about how the stencil buffer works:

80 LEVEL

29,287 просмотров • 11 месяцев назад

This is a common C bug. You write a MAX macro, then call MAX(m++, n++). The preprocessor pastes the text, so it becomes (m++ > n++) ? m++ : n++. The smaller one increments once. The larger one increments twice. Stanford CS107, lecture 12
1:04

This is a common C bug. You write a MAX macro, then call MAX(m++, n++). The preprocessor pastes the text, so it becomes (m++ > n++) ? m++ : n++. The smaller one increments once. The larger one increments twice. Stanford CS107, lecture 12

tetsuo

29,513 просмотров • 25 дней назад

** 3D Engine Update 6 ** New Buffer System. 15-20% Faster !! Using Planes A and B interleaved horizontally we can get the best of all worlds with a near continuous frame buffer for fast Triangle & Quad writes and still use DMA for fast sending of the RAM frame buffer to VRAM. Normally you get one benefit without the other leading to tradeoffs. There is a caveat ofcourse, I lose a spare plane , however thats not a large concern for what im trying to achieve, i still have the full sprite load if i need to simulate ground etc - we can also multiplex sprites if needed. So each plane is full of half fill tiles ( 4 pixels set in each tile row ) , which are then slightly offset to make the full image. Also changed to H40 mode for faster DMA albiet centered view , plenty of Hud options for side panels etc. What is quite cool is the layout in VRAM is very similar to what you see on screen just with some striping . VRAM layout to the right. Theres no more VRAM used verse a normal double buffer , the trick is here to send the buffer in one frame always as theres no room for a double buffer in vram. I have to implement partial ofscreen drawing yet hence the snapping from front to back you see still . #SGDK #SegaGenesis #SegaMegadrive
0:27

** 3D Engine Update 6 ** New Buffer System. 15-20% Faster !! Using Planes A and B interleaved horizontally we can get the best of all worlds with a near continuous frame buffer for fast Triangle & Quad writes and still use DMA for fast sending of the RAM frame buffer to VRAM. Normally you get one benefit without the other leading to tradeoffs. There is a caveat ofcourse, I lose a spare plane , however thats not a large concern for what im trying to achieve, i still have the full sprite load if i need to simulate ground etc - we can also multiplex sprites if needed. So each plane is full of half fill tiles ( 4 pixels set in each tile row ) , which are then slightly offset to make the full image. Also changed to H40 mode for faster DMA albiet centered view , plenty of Hud options for side panels etc. What is quite cool is the layout in VRAM is very similar to what you see on screen just with some striping . VRAM layout to the right. Theres no more VRAM used verse a normal double buffer , the trick is here to send the buffer in one frame always as theres no room for a double buffer in vram. I have to implement partial ofscreen drawing yet hence the snapping from front to back you see still . #SGDK #SegaGenesis #SegaMegadrive

Shannon Birt

12,758 просмотров • 4 месяцев назад

A bit late in the day, but lets build a mini-chatgpt in a few lines of Python code👩🏾‍💻🐍✨ We'll use langchain-azure-ai to access Github models (which lets us build with llms for free), and then create a chat interface with streamlit💜 Did it it in *almost* 5mins😂 Code below!
7:31

A bit late in the day, but lets build a mini-chatgpt in a few lines of Python code👩🏾‍💻🐍✨ We'll use langchain-azure-ai to access Github models (which lets us build with llms for free), and then create a chat interface with streamlit💜 Did it it in *almost* 5mins😂 Code below!

Marlene Mhangami

30,590 просмотров • 1 год назад

Pausing to decompress in a parked car can help you reset, if you keep a few things in mind. That temporary pause can act as a buffer between one part of the day and the next.
1:13

Pausing to decompress in a parked car can help you reset, if you keep a few things in mind. That temporary pause can act as a buffer between one part of the day and the next.

The Associated Press

86,697 просмотров • 4 месяцев назад

🚨 Breaking: The NYC Council voted to pass the synagogue protection bill with a veto-proof majority of 44 of 51 members. The bill to create a buffer zone for protests outside educational centers also passed, but only by a 30–19 margin — Mamdani could veto it if he chooses.
0:44

🚨 Breaking: The NYC Council voted to pass the synagogue protection bill with a veto-proof majority of 44 of 51 members. The bill to create a buffer zone for protests outside educational centers also passed, but only by a 30–19 margin — Mamdani could veto it if he chooses.

Jacob N. Kornbluh

551,980 просмотров • 5 месяцев назад

🇮🇱🇱🇧 Netanyahu: "Trump plans to invite me and the President of Lebanon to move this [peace] agreement forward. I agreed to a temporary ceasefire of 10 days to try to advance this agreement. We remain in Lebanon in an extensive security buffer zone. This is a 10 kilometer security buffer zone. We are not leaving." Source: i24NEWS English
3:07

🇮🇱🇱🇧 Netanyahu: "Trump plans to invite me and the President of Lebanon to move this [peace] agreement forward. I agreed to a temporary ceasefire of 10 days to try to advance this agreement. We remain in Lebanon in an extensive security buffer zone. This is a 10 kilometer security buffer zone. We are not leaving." Source: i24NEWS English

Mario Nawfal

115,078 просмотров • 4 месяцев назад

Theo using FFmpeg to build SoraMarker is akin to using create-react-app for a new project in 2025. On the web, it's slow, unnecessarily bulky, and makes very little use of the platform. I rebuilt it using Mediabunny, and it runs 5x as fast while using 0.8% the bundle size⚡️
0:29

Theo using FFmpeg to build SoraMarker is akin to using create-react-app for a new project in 2025. On the web, it's slow, unnecessarily bulky, and makes very little use of the platform. I rebuilt it using Mediabunny, and it runs 5x as fast while using 0.8% the bundle size⚡️

Vanilagy

46,913 просмотров • 10 месяцев назад

Anthropic pays $750,000+ a year for engineers who can build LLMs from scratch. Not how to prompt them. Not how to fine-tune them. Not how to build RAG pipelines. But how to build them from scratch. This 2-hour Stanford lecture teaches you everything. Scaling laws. Data collection. Architecture design. Post-training alignment. Free. From Stanford. Watch first. Then read this. The lecture is the theory. And this article shows you how to actually build it (with code) ↓
1:44:23

Anthropic pays $750,000+ a year for engineers who can build LLMs from scratch. Not how to prompt them. Not how to fine-tune them. Not how to build RAG pipelines. But how to build them from scratch. This 2-hour Stanford lecture teaches you everything. Scaling laws. Data collection. Architecture design. Post-training alignment. Free. From Stanford. Watch first. Then read this. The lecture is the theory. And this article shows you how to actually build it (with code) ↓

Rahul

423,606 просмотров • 2 месяцев назад

The Razer Prio is a highly portable controller for your phone that can fold upon itself to the size of a deck of playing cards. It weighs about the same too.
0:57

The Razer Prio is a highly portable controller for your phone that can fold upon itself to the size of a deck of playing cards. It weighs about the same too.

GameSpot

63,087 просмотров • 9 дней назад

Our Android Instant App is here! Yep, they're not dead 😅. Share your location for 2 hours and let the receiver see the magic ✨. It was a blast to build (with a few headaches). We did it in a few days thanks to our modular codebase, but struggled a bit with the app size (~3MB).
0:31

Our Android Instant App is here! Yep, they're not dead 😅. Share your location for 2 hours and let the receiver see the magic ✨. It was a blast to build (with a few headaches). We did it in a few days thanks to our modular codebase, but struggled a bit with the app size (~3MB).

Cyril Mottier

16,187 просмотров • 1 год назад

350-pound “plus-size model” records herself harassing a Target cashier over a XXXL bikini costing more than a size L before the cashier attempts to explain that the larger size requires more fabric, then ultimately relents and gives her a discount.
0:35

350-pound “plus-size model” records herself harassing a Target cashier over a XXXL bikini costing more than a size L before the cashier attempts to explain that the larger size requires more fabric, then ultimately relents and gives her a discount.

Right Angle News Network

60,921 просмотров • 1 месяц назад

A chipmunk cheek can stretch up to three times the size of the chipmunk's head and hold up to 12 acorns. 🐿️🌰
0:22

A chipmunk cheek can stretch up to three times the size of the chipmunk's head and hold up to 12 acorns. 🐿️🌰

Fun_Facts

52,291 просмотров • 1 год назад

Anthropic pays $750,000+ a year for engineers who can build LLM architectures from scratch. Stanford taught the entire thing in 1 hour lecture & released it for free. Bookmark & watch this today before someone takes it down.
1:29:02

Anthropic pays $750,000+ a year for engineers who can build LLM architectures from scratch. Stanford taught the entire thing in 1 hour lecture & released it for free. Bookmark & watch this today before someone takes it down.

Roan

2,591,412 просмотров • 4 месяцев назад