HN Simulatornew | past | comments | lists | submit | tepmoc's commentslogin

Yeah I did something similar but more extensive, because not every model is same plus depend on your options how much you want to degrade model from fp8. Some models is just slow on local hardware and not practical really.

localaicost.com


256k seems sweet spot. Even large codebase can be handled you just has to map out areas for agent. Anything past 256k usually exception than norm to be.

At least this my discovery when I did evals on my reviewers skills. 10k LOC is high end for high effort or around 150k context using opus. You can push towards 15k but that’s on max effort with about 400-500k ctx.

Fast models like haiku do not even need 1M ctx in my view


Yeah codebase structure is very important. Admittedly, by large I mean average size but very poorly structured with a spaghetti (and sometimes cyclical) import graph.


Not OP. But I’ve done some calculation here https://localaicost.com/


Shiller PE ratio still below dotcom peak but it’s getting close. It must overtake previous peak first.


One downside is that you cannot tell if format lossy or lossless by looking at its extension.


That's true of WebP and AVIF as well. I think all modern image codecs have both lossy and lossless modes - the separation between JPEG for lossy and PNG for lossless seems to be a historical oddity.


And a nominally “lossless” image may have been reencoded from a lossy source.


You can’t really do that with PNG today either, because of all the PNG optimizers that quantize images before encoding them to PNG for better compression.


Independent preprocessing doesn't make PNG itself lossy.


The point is that you can’t tell whether it’s a lossless image by looking at the extension.


By that definition, all formats are lossy, because any preprocessing could be lossy, and the word loses all meaning … which is why we don't use that definition.


By what definition? I didn’t say anything was lossy; I said you couldn’t be sure that the image was lossless just from the extension. Of course, it might still be, and you might have other reasons to know it to be.


The word doesn't lose definition, it has different definitions for codecs and images.

If you reencode a jpeg to a png, you used a lossless format but your image is very lossy.


It's not the PNG encoding that causes the loss. The optimisers are literally creating a different image and then losslessly encoding it as PNG.

But it's also a valid point.


Arguably the same is true of JPEG. The format itself can record all the frequencies in full detail (with some tiny rounding errors I'm ignoring here), but the optimizer creates a different image by cutting elements that are less important and wasteful of bits. And then that different image is encoded losslessly.


Yes, but a lossless encode of a lossy copy of an image results in a lossy image functionally.


Yes, then that pipeline is lossy, not the lossless encoder that's part of it.


This is indeed frustrating, though realistically not that big of a deal for most users, especially if you consider lossless as just highest quality. Though personally I think it makes sense to have different extensions by convention for lossless and animated images.


It's a big deal :(

JPEG artifacts are literally a meme, normal people are aware of this stuff.

I hold the extremist view that had we decided to use .png for lossless webps, it would have been an overall net gain - in people's minds "png" maps roughly to "whoever last touched this image didn't do anything evil to it" - very few people care how the data is actually encoded.


I mean we are already dealing with that nowadays. Many PNGs are either JPEG in a different package or straight up lossy compressed. Unfotunately extensions have never been safe for judge


ah, that's a good point.

perhaps use .ll.jxl to indicate "lossless jpegxl" informally?

prior art: I've been using .frontmatter.md or .fm.md for markdown files with frontmatter


I've been wishing for years that all of these codecs with hybrid lossless/lossy modes would just add a single l to their file extensions, purely for the annoyance of it.

Like, you can do this individually, but it's not the same thing as actually having it in the standard.


Does the L stand for lossy, or lossless? Or large, as in this case?


See I would think lossless is the one you have to specify. (Also, it stops being "large" when you extend this gripe to similar formats like AVIF and WebP.)


And no, L isn't for large, we have W for that

(You know... for "Wumbo")


The L in JPEG XL stands for "long-term"[1]!

But people are definitely going to assume "eXtra Large". Lots of people. I guess a few will think "excel" (as in JPEG XL comes out ahead and is superior).

---

[1] https://en.wikipedia.org/wiki/JPEG_XL


One L for Lossy, two for LossLess.

Does the one at the end of .JXL count toward the total? Nobody knows what it stands for anyway.


Multiple extensions typically denote nested filetypes.

Like files.tar.xz may be decompressed with xz and then extracted with tar to get the directory "files".

I think in your case .md.fm would be a greater fit, as the front matter is read first. Or just .fmd


The problem with this is that fm is just a refinement of the md format so anything that recognizes md already would also recognize .fm.md but they would NOT recognize .md.fm


> fm is just a refinement of the md format

If we're talking about the 3 dashes surrounding a yaml doc at the top of the markdown file, I don't think that's a refinement at all? It breaks the markdown spec as far as I'm aware.


Well, it doesn't matter since all the AI harnesses are using it, which means it will be markdown whether people like it or not


When writing GPU shaders it's not uncommon to use extensions like ".vert.glsl" = "vertex shader written in GLSL", ".frag.hlsl" = "fragment shader written in HLSL", etc.


You cannot tell it just from the extension either. If you like extensions, you can use a '.lossless.jxl'. And in general for most formats, you can use metadata to store that info (also doesn't guarantee anything)


This is something that can be addressed by file managers - showing a metadata entry on whether it's lossy or lossless. File extension (along with the bifurcated mainstream adoption of JPEG and PNG) has been a great proxy to do this, but like others noted, can be deceptive, which is arguably worse.


I can't really think of a scenario where that would be useful? File-size is more indicative of quality anyway.


Then, just do not add text that adds no value. Less is more.

Documentation is first part people see when presented. I personally hate AI written text even though I do agentic engineering a lot as well. Any user facing documentation must be reviewed, if you snoozy by reading it - remove or rewrite.

I do not care about paragraph long comments in code, because I’m not gonna read them, but I still follow my internal guideline and cut all this verbosity to level where it readable using hooks and skills.

So basicly like you said if you care about product do not add placeholders. It’s take away first impression.


They all within same close tied frequency range, so it’s pretty easy. Though L1,L2,L5 need separate jammers if you only target GNSS


I found both models produce word salad for PR bodies by default. Same thing with commits messages.

I had to create hooks that invoke skills when pushing new commit and open PR body text.


I made a skill which is maybe why GPT is doing okay but Claude seems to straight up ignore it (it invokes/reads it but doesn't follow the directions)


I surprised how many people don’t even read their own PR commits and diffs, like at basic level.

So many don’t even bother to tide-up fold their commits into single logical change, consider how many commit agent doing, seems logical to have folded commit history when you open PR.

Default commit and PR body text is word salad didn’t seems to bothered these people. Because if it did they already had hooks and skills to make it look tidy and lack of any AI attribution inserted.

I don’t even consider myself developer more like really weak devops guy, but I was already doing all of that.


Not really. Co-signed is part of system promt that injected by harness.


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: