When is Blitz to be updated?
Blitz3D Forums/Blitz3D Programming/When is Blitz to be updated?
| ||
| Newish to Blitz3D, having checked out the demo a couple of years ago and buying it a few days ago for a project, I've got to ask when is Blitz going to be updated to support modern techs? Why's it still tied to DX 7 when we've gone through DX8, 8.1 and 9.0? Where's the pixel and vertex shaders, hardware stencil shadows, etc.? Blitz is very convenient and powerful but it seems very behind-the-times, and I'd like to know what/when features will be updated, if anyone's got any news. |
| ||
| 99% chance that it won't be updated. Edit: Correction, 99% chance it won't be updated to support the 'modern' DirectX API(s). |
| ||
| actually blitz supports pixel and vertex shaders... |
| ||
| actually blitz supports pixel and vertex shaders... Sure. You just can't use them. Its support for normal mapping is mediocre at best, and cube mapping isn't as useful as it sounds. |
| ||
| Not gonna be updated?! Is it a 'dead' language then, or are we waiting for BlitzMax to introduce new techs? |
| ||
| SimonA is trying to figure out the problems with the collisions in Blitz3D. So it seems there is at least some force that is thinking about updating. Tracer |
| ||
| As stated elsewhere Blitz3D is not dead and will continue to receive updates and support well after BlitzMax is released including new features that are available under DX7 but currently not supported in Blitz3D. DX7 may seem dated but I'm sure HalfLife2 will be a good example of just how much scope there still is for delivering stunning stuff that runs on traditional 3D graphics hardware. Tracer: heh, I was wondering if anyone would notice those bumps. |
| ||
| cube mapping isn't as useful as it sounds. Correct, but if youre a good coder, they can be done realtime. (i have something in the codearchives todo that.) 'Ere, whats all this about blitz3d collision problems? i dont get any problems... Are we due to receive a two way moving object collision check that works? |
| ||
| Not gonna be updated?! Is it a 'dead' language then, or are we waiting for BlitzMax to introduce new techs? It isn't dead. If you write something with it, and compile it, it still works ;)Well, it's strange that some people bought a DX7 engine / programming language, and they expect it to be updated to DX8, DX9, etc. Some small game development studios work in a very different way: they first think about the game they want to make, then search for an appropiate engine (looking for its current features, instead of hoping that some of the thing it lacks will be added in a future update), buy it and stand with the version they bought, withut updating it (to ensure that nothing crashes with the update). This happens with some studios which use the Torque Engine. They got a stable version of the engine source code, and extend it themselves. That's what I am doing with the Cipher engine. In Blitz, you cannot extend the source code to add new things, but in its current state, it's perfect to make shareware games. You don't have shaders in it, but a stable engine perfectly capable of creating a sell-able game, like it has been proven with dozens of published shareware titles :) |
| ||
| Cube mapping can be easily done realtime, but it not very efficient. Try to make realtime water with a very large distance; no way Blitz can pull that off (I'm comparing to e.g far cry). Pixel shaders are essential, but as Jedive says; you just can't expect an upgrade to DirectX9, the whole Blitzmax development is being done as a preparation to a more flexible, extendible environment. |
| ||
| you just can't expect an upgrade to DirectX9 Oh, and the often mentioned 'backwards compatibility' bit which could be solved by- get this- simply not upgrading! But I figure that at this point in time, Blitz3D's source code is the equivalent of spaghetti. Fixing one thing would probably break another. |
| ||
| Skid, yeah, i noticed :) I've been waiting for a good update to the Blitz3D collision system :) So, any 'movement' in a thread by a BR person gets my attention :) In this case you were the last to post.. so, i saw "collision" + "skidracer" .. had to check ;) Tracer |
| ||
| Jedive : I agree in many respects, and I bought Blitz as it's perfect for what I want in my current project (with Tokamak). I don't need anything particularly fancy. My concern really is whether Blitz will be able to grow with the user (me) to provide more funcitonality when I want it, or will I have to find and learn another language? For me, pixel and vertex shaders would be a very valuable addition, providing high-quality visual effects for simple geometry scenes - the professional eye-candy without the professional modelling requirements!! |
| ||
| But, good models in blitz 3d can make your game look alot better ;) |
| ||
| 'Ere, whats all this about blitz3d collision problems? See the Blitz3D Collision Failure thread. |
| ||
| I only miss one thing in Blitz3D: Support for DXTC ( DDS ) textures to have an acceptable quality on my textures ... this technique is 6 years old now and it would be really nice to have it as a special type of texture. It does not need to be handled as regular texture, I know that it wouldn't be possible to modify this texture afterwards ... I just want to have 1024x1024 textures that don't take 16mb+ in VRAM! they would only have 3-4mb with dds support :( |
| ||
| But, good models in blitz 3d can make your game look alot better ;) You can't model spheres or floors to look interesting, but clever shader usage can create interesting texture and lighting effects. |
| ||
| ah yes! i have indeed had this before! cheers for pointing that link.. so it WASNT my code failing ^.^ i got around it by constantly starting the wheels a distance off the floor,reseting, and forcing them down to their position... if a wheel fell through, it would next frame go back up again. but it still felt unreliable :( |
| ||
| @Shifty Geezer: Instead of a big update, Mark Sibly will release a new engine module for his upcoming language BlitzMax. It will be cased on OpenGL, and it is supposed to have the latest tech-goodies, like shaders, although the best tihng for me is that BlitzMax will be multi-platform (Windows, Linux, MacOSX). This won't be free, but I think that's logical: you paid for a Basic programming language for Windows with a bultin DirectX7-based 3D engine. |
| ||
| BR's attitude to Blitz3d is - if it ain't broke don't fix it - I would only expect to see very small and token additions/fixes for Blitz3d just to keep the vultures away. In any case Blitz3d will never be updated to anything beyond dx7. While BlitzMax development is hogging the processor I doubt wether Blitz3d will get even a couple of clock cycles dedicated to it. For a couple of years now the stock answer to feature requests or bug fixes has been - BlitzMax ! I had no problem with this given that we were informed again & again that BlitzMax would have/do most things that Blitz3d lacked What was then very dissapointing was the news that BlitzMax will have NO built-in 3d commands/3d module at release and you would have to: a)Write your own. (defeats the object of purchasing a high level language in the first place) b)Use a 3rd party module (which unless is used by everyone will cause havoc when exchanging code - reliabilty would also be an issue. I can see that there may be loads of different versions etc). c) Wait an indefinite period of time before BR writes a native 3d module. Given the speed at which BlitzMax development has taken place so far (understandable given the size of the programming team - one person) - this could be a long, long wait. Many new users will look and see Blitz3d as an old and outdated language and they are right - it is old and outdated but its stable, versatile and you can still make good shareware games/tools with it. |
| ||
| So that's BlitzMax huh? I read something about it in the newsletter but now that explains what he meant bny wanting to get back into 3D. Max will be a language without any 3D support at first? I guess it's main usefulness will be as a portable basic, but I can't see existing Blitz users flocking to it and can't understand where the new customers would come from. What's BlitzMax aimed to do if not create an easy 3D programming environment? |
| ||
| Greetings Puppies, High and (seemingly) low-level multi-platform development tool, I think? Peace, Jes |
| ||
| What's BlitzMax aimed to do if not create an easy 3D programming environment? Well the short term result will be to produce a more complicated 2D programming environment. However, we are told that BlitzMax will be a more advanced language a few abstraction levels beyond the current procedural basic. From the very small code snippets I have seen It looks like it will no longer be 'basic' but more of a hybrid 'c' type language. As well as being cross-platform and more object-orientated it will remove the dependance on DirectX - the 2d graphic commands use OpenGL . In theory it should be possible to use any 2d/3d graphics library at all. |
| ||
| So, you people who are complaining about BlitzMax not having 3D support (from Mark) at release, are you saying that Mark shouldn't release it until it does? |
| ||
| What he said. I will never understand all this whining. but I can't see existing Blitz users flocking to it You're clearly new here, so what do you know of these existing Blitz users? Blitz3D is not dead, however it's development is on hiatus it seems - until BlitzMax is finished at least - and when somebody does turn their attention back to B3D it will probably mostly be bug fixes (and the few remaining things that DX7 supports that B3D doesn't) - however, as it stands, right now, the language is at it's peak of 'featureset vs stability', so it really never has been a better time to churn something out with it. Pixel shaders does not a game make - besides, think of your target audience. Do you have the manpower and resources available to you to take on the big AAA titles? Even if you managed to get a AAA class title finished how would you get it sold? It's nigh on impossible getting a major publishing house to recognise you and without heavy promotional funding you're never going to get the attention of the mass 'gamer lamer' market. Having said all that, I must stress it's not impossible either, before a whole buncha people start yelling at me =] Still, there are a number of people who have done pretty damn well for themselves on their Blitz-made games. The shareware mass are ripe for the picking and are generally more open minded, which basically means you can do quite well with something a bit less detailed but typically more interesting than your usual FPS romp - tho now you have to keep in mind that most shareware gamers have a considerably lower spec machine than your average gamer lamer. Pixel shaders probably wouldn't be an acceptable way to enhance the quality of your game's graphics anyway. So, I managed to bring my story back to the original point in the end. What a fascinating ramble. Max will be a language without any 3D support at first? Max at it's initial release will contain a complete 2d-in-3d module, similar to the current implementation, but of course OpenGL based and have all the features you'd expect from 3d based commands. There will be no 3d -engine- module at it's initial release, however the entire OpenGL standard featuresets will be completely open and accessible to the user. What's BlitzMax aimed to do if not create an easy 3D programming environment? To create an easy programming environment. can't understand where the new customers would come from. Probably the same places they're coming from now. Mostly the DBP forums =] BlitzMax is going to be the schiznitz of high level programming languages, even touted in unofficial PR by a Blitz Research member as 'the bees knees'. It will have -the same- accessibility of a high level basic language as current Blitz products do now, with the ability to extend your knowledge over time to a much lower level if you need, without losing many of the high level benefits. The capability behind BlitzMax is literally endless, but that of course relies on the strengths of it's community and/or your own programming skills. Well the short term result will be to produce a more complicated 2D programming environment. Not a more complicated 2d programming environment, but a richer programming environment. From the very small code snippets I have seen It looks like it will no longer be 'basic' but more of a hybrid 'c' type language. It will be the same BASIC-based language you know and love today, with a few minor syntax alterations and a truck load of extra useful stuff you can do if you so wish. |
| ||
| I like structured programming, and a move to structure programming in Max would be a plus for me. But for C like structuring and OpenGL support on multiple platforms, why not use Java? For me, the reason I don't use C++ and it's OO structure to write 3D games is due to all the hassle creating the models, needing to work out 4d matrix calculations and quaternion nonsense. Too much like hard work!! Blit3D gives me easy commands, createenetity, turnentity, that removes all this and offers a very fast development. It's lacking structure that I'd like, though is reasonable enough with it's implementation of functions that I can work with it fairly effectively. What I'd expect from Max is more structure with classes and methods etc., and the nice 3D functions so I can still use createentity, turnentity without having to bash my brains out with pointers and 4 dimensional imaginary numbers :) Without those nice 3D functions my development will be far easier sticking to Blitz3D than going over to Max. |
| ||
| An official 3d module is not so important to me, nor is cross-platform compilation. I just want a modern, stable compiler that produces faster executables than Blitz3d. It seems that if 3d is your thing then you will be able to get down and dirty with the polygons as soon as soon as it is released due to the wide variety of 3d engines that are already available on the net. If they are too complicated and low-level then VividGL is just around the corner too, and I don't think it will be much more difficult to use than the 3d command set that good old Blitz3d currently uses. I am planning to use BlitzMax and VividGL, but I must confess to being a little concerned about the inevitable bugs and stability issues that will arise from using these two new products together. As far as Blitz3d goes, it is already well suited to making all manner of 3d games and doesn't really need many updates other than the occasional fix for the few remaining bugs. It's time to put the old horse out to pasture. The future is Cobol!! Err, I mean BlitzMax. |
| ||
| Blitz3D is a receiving abyssmal support from the dev and has been for quite some time now. B3D is based around an API that was released what... about 5 years ago? Darkbasic Pro on the other hand has all the eye candy features of DX9 and is updated on a continual basis. I like Blitz structure better and I think the IDE is much better IMHO and DBPro IS quirky but unlike Blitz3D, it not been virtually abandoned by it's devs. So if you are looking for a basic language that supports the latest DX9 stuff then you are pretty much wasting your time here and your needs may be better served by DBPro at this point in time. |
| ||
| I wouldnt like to think Blitz3D as 'dead'. It still has potential for me. Maybe, when I have mastered all of it's finer points and have stretched its boundaries to unprecedented levels, and taken over the world with it, then I will deem it - non-viable. Until then, It is alive as yoghurt :) |
| ||
| So many people here really like their fancy effects which are all very nice and dandy but won't put any meat on the bones of a game, just a little herb and sage in my opinion....collisions aside I think Blitz is still one heck of a performer and am actually dreading the addition of more DX glitz, which will only lead to more distraction except for the really strong of will, um... in my opinion that is. |
| ||
| I agree like others in here. blitz basic ain't dead. i meak it alive and kicking in my puter, and happy it is... but but but... yes i would like the DX9 featurs, would be nice if they updated it, but okay it's a DX7 platformi can live with that if blitz MAX would ever come.. hehe. What happens with blitzMax, when will it be realised.. ?... it's like a ghost, when is the deadline (hope i just missed out on some website with infor on that.) you guy's know Processing from ben fry and casy rees, check it otherwise, it's super super simple and nice,,, well it misses a loat of things, but i love it's oop (well it's JAVA).... so what about it.. when is blitzMax to be out :) looking forward to it... |
| ||
| when is the deadline The projected estimate ran out a few months ago. I think, more recently, a new projected estimate was scheduled for 'maybe in a few weeks', which is yet to come. Personally, I'm still hoping for XMas. |
| ||
| well, whadya know, it got updated! |
| ||
| Blitz3D is a receiving abyssmal support from the dev and has been for quite some time now. B3D is based around an API that was released what... about 5 years ago? Darkbasic Pro on the other hand has all the eye candy features of DX9 and is updated on a continual basis. I like Blitz structure better and I think the IDE is much better IMHO and DBPro IS quirky but unlike Blitz3D, it not been virtually abandoned by it's devs. So if you are looking for a basic language that supports the latest DX9 stuff then you are pretty much wasting your time here and your needs may be better served by DBPro at this point in time. Not everyone wants Blitz3D to use DX9, you know. DX9 is a waste of time for shareware developers. I would prefer it if B3D went down to DX6 or DX5, although I'm happy with DX7. |
| ||
| DBpro? If you like it great, but for me, Ill take stability over quirky anyday. Id be really interested in knowing if most users of Blitz3D and DBPro really had any idea of what advantage, if any, that DirectX9 would bring to a game coded in either one? Yea, yea, pixle shaders, thats all you ever hear about, but as far as I know, these would only be useful if your planning on releasing something a bit more, as someone else called it, AAA than 99.9% of the normal BB3D/DBpro community would be capable of delivering. Lets face it, out of all the people who complain about this short coming in BB3D less than 1% of them would actually have a clue what to do with the added features. I think for the masses, BB3D has far more to offer than most will ever use. Why don't you spend more time using what you have, and less time complaining about what you dont have and would probably never use? If you did that we would have even more great games. :) |
| ||
| I downloaded the DBPro demo the other day just to see what it was like. I actually tried DB before blitz. (Actually I ran Blitz Basic on my Amiga b4 DB, so I feel better now):-) The commands are rediculously long and complicated!! Each entity is stored in an object number. This is easy enuf to give a handle though, you just give a variable the object number. But the biggest surprise was no tweening options. It has frame limiting, but thats all!! What the...? There are crappy demos, crappy docs... BUT I want to try converting my current game to it just to see how the final output compares to Blitz. If I pull it off I will post both versions for you guys to compare. What this has to do with Blitz Updates...dunno SO When's 1.89 Due??? Just kidding !! |
| ||
| "I want to try converting my current game to it just to see how the final output compares to Blitz. " just remember... the graphics and engine features are not the only difference between B3D and DBP. i personally wouldn't mind DBP as much if the language itself were better. as it stands, DBP's language is still very much like DB1 - the type support is minimal at best (and buggy in more complicated expressions), forcing you to rely on things like arrays and custom memblock data structures. the way types are set up, there are no type pointers - so you can't pass types to functions byref, you can't make your own linked list, and the only way to "emulate" type pointers is by storing types in an array and using their indices - but then that becomes a problem when you start deleting random items from the array, invalidating those indices - so instead you have to have TWO arrays, one that is for the types, and one to keep track of which elements in ths first array (if any) are open.. just a really, really big pain in the ass that could be very easily avoided if they used type pointers instead. additionally the resource index system is just as much a hindrance as the lack of type pointers - you WILL find yourself cursing when you have to work around those damned indices. the funniest thing - many people who use DBP have functions like: function GetFreeObject()
for i=1 to 65535
if object exist(i)=0 then exitfunction i
next i
endfunction 0so that they can emulate the "handle" system blitz uses. and yet they wouldn't dare change the index system. aside from that.. the engine is pretty hot, but don't believe ALL the hype - BSPs still have bugs; the new advanced terrains still aren't as flexible as blitz terrains (can't modify them in real time?!); multitexturing (something that i take very much for granted in blitz) is only offered through a handful of underdocumented and restrictive commands; effects by their very nature don't always slot into the engine very well (and no animating models with effects.. boom); the inbuilt shadow shading is pretty crappy (on a lot of cards, the shadow is cast, rather jaggedly, onto the backfaces of the model that's casting the shadow); and the collision.... is nonexistent. you need NGC to do ANY kind of worthwhile collision in DBP. oh, and the EXE sizes get larger and larger with every release. would you believe that the entire DBP runtime is over 11MB with the latest patch? sure it's hard to get ALL the DLLs included, but the EXEs are averaging about 5 or 6MB minimum with this patch. |
   |