Anyway, here's a preview of the server side of my protocol code. It will support big maps but it won't be compatible with the FitzQuake protocol because the original algorithm of SV_WriteClientdataToEntity is a mess. I've only made an effort to keep protocol 15 compatibility because the original startdemos uses it.

https://drive.google.com/file/d/0BxGzcs7XcizgOW1zS0ZESmd6UGs/view?usp=docslist_api 

Retroquad is purely software rendered, no GL whatsoever? Are you making those shadows and bloom without OpenGL? 

And I'm getting better at it.

In the past I wanted to write a hardware-accelerated renderer, but I don't think that's necessary anymore.

It's a shame that Carmack abandoned his own software renderer, because the ideas behind it still have a lot of potential. The more I master it, the more things I realize that can be efficiently done.

The error most people makes is to try to mimic hardware acceleration verbatim. I'm ignoring that and focusing on the software renderer's strengths. 

So, about the description: it seems that you're speaking to yourself (a lot, by the way), not to your user. The description doesn't make crystal clear what makes RQ stands out, where's its place among other ports. Is it the "software render" guy? Is it the "console Quake" one? The "modular code port"? What's its motto? The description strikes me more like a big "thinking out loud" and less like a pitch for dummies looking for an engine like me.

Hope these 2c are helpful someway. 

Yeah you need to cut down on the ramble. To be honest you could probably cut down the text to a quarter of its size without losing the salient points.

Like this is a typical paragraph, and most of it is just waffle to be honest:

"To make sure that it all comes out as expected, with everything working well not just in the technical sense, but also in the artistic sense, one of the main guidelines I've established for this engine is "feature consistency". It means that if a certain feature is available for a certain kind of thing, it should also be available in some similar form to all similar kinds of things (e.g. in Quake, only static surfaces, MDL models and internal BSP models were lit, but in Retroquad, everything else - external BSP models, liquid surfaces, skies, 2D sprites and particles - can also be lit), and it should adapt itself to behave consistently in any user configuration (no matter the screen resolution, aspect, FOV, and so on)."

I would just turn that whole lot into a heading and a paragraph focusing on the important bit, e.g.

Feature consistency
e.g. in Quake, only static surfaces, MDL models and internal BSP models were lit, but in Retroquad, everything else - external BSP models, liquid surfaces, skies, 2D sprites and particles - can also be lit. 

First, and most importantly: It all seems as if you're asking me to pay for your hobby. I do understand that your financial situation is difficult, but I still don't really see why I should give you this money.

Second, I don't know who your target audience is. Is it players? They are not interested in all those technical descriptions I'm afraid. I suppose you're really trying to market this engine to modders and indie game developers, but even then, it remains unclear why they should choose your engine over others.

Finally, if you do target indie game developers, then you need to add a roadmap detailing when certain features will be finished and when you expect the entire project to be done. Your approach seems to be "I'll just release it all when it's done, with some files here and there when I feel like it.", but seriously, who is going to wait for that and be content to shell out their money on trust alone? I certainly would not.

To be truly honest, all this seems like you're trying to make some money off of your hobby, which in principle is fine, but in this case I think it's hopeless. Your engine is way too niche to be interesting to game developers when there are plenty of free, and already finished, alternatives out there which do run on more platforms and with a proven track record. They few things that set your engine apart don't justify the money nor the time a dev would have to wait for you to be done (which might very well be never).

I'm sorry if this sounds harsh, but it is my opinion :-( 

Hmm, let's see.

"The description doesn't make crystal clear what makes RQ stands out, where's its place among other ports. [�] What's its motto?"

Refinement.

All engines focus on improvements, but refinement is a more strict approach. Sorry, I don't know how to explain in a more concise way.

Unifying the features of all official ports is part of this refinement.

"To be honest you could probably cut down the text to a quarter of its size without losing the salient points."

That's my weakness. I'm not good resuming stuff, which is why I've asked for some opinions.

But you've cut a bit too much. The last bit is also important, because I'm not aiming just for the presence of the features. I'm also aiming to make their behavior consistent, and that's important because the vanilla software renderer behave in awfully inconsistent ways (particles becomes _smaller_ when you zoom in, for example). I just don't know how to make that clear in a more concise way. 

Those are valid points.

Yes, the target audience is also indie developers.

"but even then, it remains unclear why they should choose your engine over others."

I'm certainly not trying to compete with Unity or UE4.

And I admit that this is what I have the most difficult to answer. If an indie dev feels at home using other engines, there's no reason to change.

But, just like there still are people developing games for retro consoles such as the Genesis or the Jaguar, there may be some people interested in developing PC games with a current engine that's truly retro. I see a lot of indie games doing this with 2D engines, but 3D indie games usually tries to fake the style of software renderers using hardware renderers, what in my opinion is just plain wrong.

An indie Genesis game running on an actual Genesis is a technical achievement; a 3D game underusing a hardware-accelerated to simulate a software renderer is not.

I guess my main audience is people interested in making that kind of technical achievement, but I admit that's a really small niche indeed.

"Finally, if you do target indie game developers, then you need to add a roadmap detailing when certain features will be finished and when you expect the entire project to be done. Your approach seems to be "I'll just release it all when it's done, with some files here and there when I feel like it." "

It's not when I feel like it; it's just that I don't like to make promises I'm not absolutely sure I can fulfill, and giving exact dates is one of those promises. My life is too unpredictable, because any small problem is taking a big portion of my resources to fix. For example, if the roof of my house falls down, I'd be desperate and unable to deliver the code in time� which would make me double desperate. That's an extreme example, but it's actually possible.

I feel a lot of remorse when I don't keep up with my promises. 

Even if you develop your engine based on Quake, would an indie dev be able to sell it? Surely zenimax would get really pissed off about it? 

It's still under the GPL, so no problem. 

Your competition is DarkPlaces, QuakeSpasm, FitzQuake, and the like. The only thing your engine has going over these appears to be the software renderer. But I also expect that using it comes at a cost of performance and platform availability (no Mac OS X, for example). I'm really not sure if the software renderer is enough of an argument to sell an indie dev on your engine.

Regarding a timeline, I understand where you're coming from, it's the same with me and TrenchBroom, but for different reasons. That's why it's free and open source. No one can, and no one does, expect me to make promises as to when a release happens, and what features it contains. You on the other hand want money, but without any commitment. That's not gonna work for a lot of people I'm afraid. It's also why it looks like you wanting others to fund your hobby projects. 

Asking people for sponsoring a hobby that leads to public products is perfectly fine and pretty much what Patreon is. It's a hobby for the receiver of the monetary support, they can justify extra time investment or feel rewarded. It's 'paying' for something nice for the patrons. 

Spirit is basically right. Mankrip is doing exactly what patreon was designed for... it's literally the reason it exists :P 

If mankrip were making a commercial-like product he wouldn't be asking for donations at Patreon.

I don't think it is fair to compare a hobby project to a commercial-like project. 

In this case I believe Strafe 1996 is communicating perfectly to your audience.

Sleepwalkr, patreon is not like crowdfunding. Though I agree there must be a clearer quid pro quo in this case.

I feel like giving some unsolicited sugestions, bare with me: maybe we need more project management here; maybe it should be a code refactoring/doc project, aiming programmers, something like Quake Code Bible. Is there onde already? 

software renderer .. (no Mac OS X, for example*cough* Mark V has a Mac software renderer. 

Though I agree there must be a clearer quid pro quo in this case

... to be more successful, to get more patreons. 

I didn't understand the nature of Patreon. But my other objections stand. 

Typically, Patreons WILL get something in return tho. In your case, maybe you could do a weekly stream or something where you work on the engine in real time and interact with people? Like Handmade Hero does... 

The last bit is also important, because I'm not aiming just for the presence of the features. I'm also aiming to make their behavior consistent, and that's important because the vanilla software renderer behave in awfully inconsistent ways (particles becomes _smaller_ when you zoom in, for example). I just don't know how to make that clear in a more concise way.

As written initially it's a bit vague and leaves people wondering what exactly you mean. Just saying features will behave consistently no matter what screen resolution you are in will just result in a "ok...so what?" reaction.

Giving an actual example of what it fixes is much better. However, details of minor bugfixes like that should probably be left for a seperate section which just lists (in concise bullet points) what it does differently to vanilla quake. 

there are some good ideas about Retroquad. The idea of a quality modern software renderer. The idea of refactoring and documenting a legendary engine's code. The idea of running this legendary game/tech on Android. Lot of potential to explore. 

You get a func post. 

The Func crowd is composed of critics, and rightfully so.

Criticism is the aspiration for excellence.

Although there is no bad "publicity", unless it is specifically criticism and high standards you are looking for, this may not be the kind of feedback you are looking for.

[Translation: when I've posted about highest standards "completed and tangible" projects here, it has gone well --- the others, not very well. I wouldn't even think of posting about a work-in-progress here unless it was very clearly known to be such *AND* I needed a mapper's point-of-view to complete it correctly.]

/Free internet advise. Probably no better or worse than other "free internet advice". Haha ;-) 

However, what's being criticized is not the engine; it's the proposal.

Anyway, if I have to justify it, that's because it was rejected.

I won't dispute this anymore, Right now I just want to write the engine, and writing about it won't help with that. I'm happy with how it's shaping up, and that's enough for me.

Thank you all for your solid reasoning. I'll keep to myself from now on. 

There is lots more to the Quake community than just the nitpicking know-it-all func people.