14849. Anonymous (125.195.55.66) –
05 Apr 01:38:11
[ THPS Rules! ]
I've been stalking you on SourceForge for a while (but tried my best not to spoil the fun), and I think it'll be quite challenging to turn your work into video this time. Please don't feel discouraged by the amount of editing work this might require! Of course I have no rights to represent your fans, but I'd still like to say that we'll fully support you, at least emotionally. Take your time, don't let this burn you out, we'll wait. At the end of the day, I really want this to become a video, and your videos are by far the best videos I ever watched on YouTube. |
| |
14848. Sean (123.16.151.44) –
02 Apr 00:28:51
[ THPS Rules! ]
You know in the original ReLU paper they were more than hinting that ReLU was a switch. But everyone latched onto the function viewpoint.
That was the wrong thing to do. ReLU either conducts 1 to 1 or outputs zero. Therefore it is a switch, with switching decision (x>=0)?
Then weighted sums are being connected to and disconnected from each other. And those connected weighted sums can be simplified by linear algebra at each stage. You can gain a very clear insight into how ReLU neural networks operate.
You can look up switchnet, or WHTeBooklet or a frozen neural network on Archive.org. Or somewhere in this booklet I talk about it: https://sites.google.com/view/algorithmshortcuts/walsh-hadamard-transform |
| |
14847. Qaziquza (108.216.110.132) –
01 Apr 17:10:30
[ THPS Rules! ]
Can't wait to see what you've done this time for SIGBOVIK! |
| |
14846. Tom 7 (166.199.151.62) –
30 Mar 17:53:40
[ Leap day! ]
Jr: Drat! I have been thinking I need to do another ldjam, too, but the overlap would be even more prohibitive for me :(
Anon: Why not? I've got good taste!
Salt: That's how I felt too.
Good day, mysterious pi ninja! |
| |
14845. Anonymous (174.229.116.185) –
14 Mar 18:49:00
[ Leap day! ]
Happy pi day tom7. From your sidewalk chalking partner. |
| |
14844. nacl (205.175.106.64) –
06 Mar 19:50:15
[ Leap day! ]
I wish I wasn't as excited as I am to see that the pin puller game actually exists, but here we are. I am going to play EVERY level of it. I am what I am. Thanks Tom. |
| |
14843. Anonymous (71.68.80.132) –
04 Mar 15:29:41
[ Leap day! ]
Did not expect to see Celeste and Unnecessary Detail here! |
| |
14842. Tom 2 (108.14.168.243) –
04 Mar 01:21:12
[ Leap day! ]
Ah man, would love to come to An Evening of Unnecessary Detail, but it overlaps with Ludum Dare! Will keep on eye out for future appearances. |
| |
Glad you liked Slay the Spire, it was so successful that I feel a lot of new indi games just try to copy it. I think it's the level of balance and refinement that makes it as good. |
| |
You should give team fight tactics a try! |
| |
I am not convinced by your proof that "generalized kerning" (which i understand to mean arbititrary finite kerning + arbititrary finite ligatures) is undecideable.
I followed your proof but i would not describe it as "generalized kerning" but as "generalized kerning with substring acceptance".
I suspect your substring extension is necessary to proof undecidedability because "generalized kerning" (without substring acceptance) is decideable. You observation that generalized kerning can implement a turing machines is correct. I suspect that "generalized kerning" (without substring acceptance) is bounded in a similar manner to how tape bounded turing machine are bounded (and decideable).
I think all ligatures and kerning examples you presented were "local enough" and seem to require less then what a context-sensitive rewriting system would offer. Which are decideable as off (2020) https://verify.rwth-aachen.de/giesl/papers/GieslMiddeldorp-distribute.pdf . I think the ability to hide an arbitrary amount of steps allows you to escape bounds of the world of linear bounded turing machine.
You would need a description of the rules of generalized kerning to get a more detailed answer. |
| |
I've had trouble with MSYS2 when compiling both POSIX and Windows programs. Or rather, one program, PForth -- Phil Burke's Portable Forth -- with separate POSIX and Win32 IO code. Compiling the POSIX code, the MSYS2 compiler errored out on some printf syntax. Compiling the Windows code failed due to a missing header. I ended up using a different MINGW distribution designed for compiling Windows programs with llvm. It lacks a shell and commands such as rm which makefiles need, so I use it by prepending it to my path in MSYS2. There are many MINGW distros; perhaps you can find one with a compiler which suits you and perhaps run it with MSYS2 as I do. I can't remember what mine is called beyond the dirname, llvm-mingw-20220906-msvcrt-x86_64 and I'm too lazy to upgrade it. |
| |
Thanks guys. I love that they built WSL, but it's probably not for me since I also want to be able to make native code that works on other people's computers (for example, our destroyfx.org plugins, which are DLLs). I could just give up and use MSVC for those, and for native game development (which I don't do that much these days anyway), but 25 years ago MSVC would crash when I used C++ templates a little too hard and I never forgave it. The "cross-compilation" is actually great for me, except when I'm trying to use someone else's software that wants like "libxml2" and that doesn't have packages for windows.
I do use virtualbox to run Ubuntu and frequently have that up for development tasks that work better on Linux. That's a pretty good setup, especially since it makes it easy to do some sandboxed experiments without worrying about messing up my "real" computer, and it's easy to reason about what's happening. I had been using it recently for whenever I needed AddressSanitizer help on a head-scratcher, which was part of the impetus to try to get that working on windows. |
| |
There is definitely a SIGBOVIK opportunity for Chat GUID Partition Table. |
| |
I can second WSL as a good "I just want a ding-dang C++ compiler on my windows PC". Especially now with WSL2 having an x11 compatibility layer so you see plots and visualizations.
But having been in the same boat, I'd just bite the bullet and dual bit with a Linux distro. In the long run it has ended up as less headache than wrestling with all the little nuances of these tools on windows |
| |
Have you experimented with WSL? I found it to be a really great way of doing Linux on Windows, but I imagine you have some very weird and specific needs that "run it in a VM that happens to be well-integrated into the OS" doesn't satisfy. |
| |
That's great! It is fun to write books and a great honor when someone reads your books :) |
| |
I haven't played it! But it looks interesting and I just added it to my wishlist. Thank you for the recommendation :) |
| |
|
|
| |
sounds like your ladder has an attitude problem |
| |
hopefully those pdf renderers out there can display your pdf files. i think that would be too easy, though. |
| |
wonderful stuff :) happy Yule |
| |
I had a fun time reading the book, it was strange and demented, and thoroughly fun. It reminded me that books don't have to make sense, which has inspired me to write my own, of a similar format. |
| |
I just remembered your novels when talking to my girlfriend about some surrealist-adjacent books the other day--getting the hardcopies right now. (By the way, looks like the Lulu links may be broken, but the books still show up when you search for them.)
I'm always inspired discovering your various projects--they tend to make me wish I had whatever job you have so that I could have the free time to be as creative as you...but apparently now I have no excuse because I just realized we are coworkers.
Thanks for sharing your enthusiasm with technology/arts/more with all of us! I have to agree with the above immortal words of Anthony Michael: truly "ROBIN WILLIAMS GENIUS" stuff. |
| |
Tom, have you played the abstract puzzle game Insight yet? That one is long, so don't start it right before some important deadline like I would do.
|
| |