Відео

These videos are JUST PURE GOLD

For those unable to see their own named general purpose register names (labels) in watches and variables, do this: Click the toolbox (Project Properties) icon from the Dashboard (extreme upper left icon), then select 'mpasm (Global Options)' and check the box for 'build in absolute mode', apply and exit. Then debug. It's taken me a couple of years to getting around to finding this fix, I have searched countless Microchip forums and watched as many YT videos to no avail. Hope this helps you.

where is the source code. please push it to your github repository

Thank you for the video! Is your website still active? I can't find it anywhere.

Compilers HATE This One Simple Trick...

thank you so much from Turkey

absolutely fantastic in content and presentation! Many thanks!

great piano.

Nice!

Hmmm. Well, I guessed correctly about Dekker, and would not have had to asked my earlier question if I had bothered to watch the whole thing before asking my earlier question. He was indeed Dijkstra’s colleague, another one of those early titans of computing science…

Sorry I’m so late to the party! Just discovered your marvelous AVX512 introduction. Is TJ Dekker the same Dekker who derived the correct mutual exclusion algorithm at the Dutch Mathematical Center in the 1950’s, when EW Dijkstra was building the THE operating system?

Another way to generate the Fibonacci numbers using XOR assignment for swapping the variables: for(int i = 0, j = 1; ; j^=i^=j^=i+=j );

In your example at 11:48, you better should put the "NotLowerCase" label 1 line down to not make a useless "mov"

Milliseconds -> ms /sarcasm

Excellent video - thank you.

I tried this example of SOA when loading 50 large textures, on my game written in C.The rest of the variables were primitives, and were initialized first. It was so fast! But I found that the loading speed was increased dramatically if I loaded the textures first. I’m talking about 5 seconds to load, to under 1 sec. I learned that positioning of variable initialization is important too 😂 It also makes the code so much easier to read.❤

I’m an undergraduate in CpE and I was working on an autonomous robot, and I had this really over engineered way to do course correction. I watched this video, and just today I was looking over my system and noticed that the techniques in your video were perfect for this application. I reduced it from a 50 line solution of ugly nested if else’s about 4 deep at max, to 2 lines. Thank you!

The best way I know to do the branchless translation from lower case to uppercase is to use a 256 bytes translation table where every byte contains its offset from begining except those from offset 'a' to 'z' containing 'A' to 'Z', so the loop's body reduces to : d[i] = tbl[d[i]; Another thing when you try to minimise branches is to remember the "hack interval" : as long as a and b are known to be non negative integers, the condition a <= x && x <= b is equivalent to (unsigned)(x-a) <= (unsigned)(b-a). In the general case this will replace 2 branches by one, when used in a loop, in most of the cases a and b don't change in the loop, so the compiler moves the calculation of (b-a) outside of the loop, here we have constants, so (b-a) is computed at compile time. Here, if we want to avoid the translation table, the body of the loop can reduce to : d[i] -= ((unsigned)(d[i] - 'a') <= (unsigned)('z' - 'a')) <<5; It should be easily vectorisable.

okay series, but great fucking drawing. Your dad totally stold the show. The detail is incredible.

I've actually never thought about branchless programming in the context of making a program faster. I only thought of it in the context of cryptographic systems where you want to ensure every execution has the same speed and heat profile or cryptographic systems which just don't allow for branching

Your accent can be mistaken (by some American like myself) for Australian, and with those trees in the background, I just imagine what you're saying to be like "And we can see these Koala bears teaching their young some integer division using coconuts. Yep, and whenever a coconut rolls off of the tree branch, it sets the carry flag to one Or perhaps it IS Australian. I was like, he sounds British, in his earlier videos, but now I definitely hear something my brain identifies as Australian or similar.

Love when experts have a down-to-earth attitude like this. You rock man.

This added some much needed context to the content in my computer architecture class, thanks!

this is why i prefer RISC

Thays crazy 😧

9:57

5:59

31:24

0:12

11:55

1:44

19:11

1:04

0:13

Branchless machines are not Turing-complete.

This is an awesome video series, thank you for it. However, how would you apply it to some of the newer ieee754 types, like _Decimal32, _Decimal64, and _Decimal128? I can see these being problematic for a couple of reasons. One, there is frequently more than one bit representation for the same number, particularly if the number of decimal digits is smaller than the maximum digits that can be represented in such a value, and two there are multiple standards for decimal floating point numbers.

Awesome Thanks. I was able to code this with nasm, Debian12 and qemu. :)

Thanks for the intuitive and incredible tutorial about bank conflict!

Crazy? The last one is more like crazy usefull and powerfull.

saying that it is worth your time to get rid of what is factually BREAD AND BUTTER of whole programming -- conditional operators -- is insanity. overoptimisation is one of worst enemies. sure video is fun, but c'mon telling it is legit programming technique in a world with hardware and compilers of curerent time? madness on modern hardware (and by modern u mean 20-25 years) it's totally not worth wasting your own time to get into such witchcraft. also, say, converting upper to lower takes 0.005% of your program processor time, so you make your program not 20 times faster, but 0.01% faster. if it works. compiler knows better, always. and processors themselves know better. prediction, caching, parallel execution all this stuff. they were good in it 20 years ago, you can imagine how good they are now. such heavy optimizations are only worth it if your profiler shows you heavy peaks somewhere. but in this case 99% if time you are doing something really wrong and can fix it by refactoring abit.

I know this is old, but I am vonfused about something... How is (a<b) not a comparison? Isnt it essentially: if a < b then return 1 else return 0?

UA-cam keeps deleting my comments :(

In the 21st century, with efficient optimising compilers and high clock speeds, why muck about with assembler? Difficult to read, write and maintain!

Thanks 🎉🎉

I implemented your C procedures, but declaring the structures and doubles volatily to prevent GCC from optimising the code, multiplied 0.1 50 times to get ~1e-50 and compared it to a primitive multiplication (p *= 0.1). The error with DekkerMul is 2.8486703237279e-065, the one with the primitive approach is the slightly different margin 2.4925865332619e-065. What did I do wrong ? I summed up R.upper and R.lower before starting the next call to DekkerMul. Thank you.

Great video.. you earned a subscriber!

The fewer conditionals in a program, the better. But we get paid for conditionals. So be careful, human!

Don't make global libs directory. I did it once, and regreded it. It seem like good idea untill you install new version of some lib, and later you notice that some of your projects don't compile. Just put all required libraries into your project directory.

Holy crap I understand why a segmentation fault is called a segmentation fault now. Thank you.

How did you design the roulette spin function? Did you program it to account for RED BLACK Odd Even 1-18 19-36? What if the players change what they are betting on each spin? This program is not accurate for predicting an actual roulette betting session. Random numbers don't account for how the roulette wheel numbers are actually on the wheel and plus the randomness of ball bouncing around and changing directions.